Kernel based subspace projection of hyperspectral images
Larsen, Rasmus; Nielsen, Allan Aasbjerg; Arngren, Morten
In hyperspectral image analysis an exploratory approach to analyse the image data is to conduct subspace projections. As linear projections often fail to capture the underlying structure of the data, we present kernel based subspace projections of PCA and Maximum Autocorrelation Factors (MAF...
Kernel based subspace projection of near infrared hyperspectral images of maize kernels
Larsen, Rasmus; Arngren, Morten; Hansen, Per Waaben
2009-01-01
In this paper we present an exploratory analysis of hyper- spectral 900-1700 nm images of maize kernels. The imaging device is a line scanning hyper spectral camera using a broadband NIR illumi- nation. In order to explore the hyperspectral data we compare a series of subspace projection methods ......- tor transform outperform the linear methods as well as kernel principal components in producing interesting projections of the data.......In this paper we present an exploratory analysis of hyper- spectral 900-1700 nm images of maize kernels. The imaging device is a line scanning hyper spectral camera using a broadband NIR illumi- nation. In order to explore the hyperspectral data we compare a series of subspace projection methods...... including principal component analysis and maximum autocorrelation factor analysis. The latter utilizes the fact that interesting phenomena in images exhibit spatial autocorrelation. However, linear projections often fail to grasp the underlying variability on the data. Therefore we propose to use so...
Binol, Hamidullah; Bal, Abdullah; Cukur, Huseyin
2015-10-01
The performance of the kernel based techniques depends on the selection of kernel parameters. That's why; suitable parameter selection is an important problem for many kernel based techniques. This article presents a novel technique to learn the kernel parameters in kernel Fukunaga-Koontz Transform based (KFKT) classifier. The proposed approach determines the appropriate values of kernel parameters through optimizing an objective function constructed based on discrimination ability of KFKT. For this purpose we have utilized differential evolution algorithm (DEA). The new technique overcomes some disadvantages such as high time consumption existing in the traditional cross-validation method, and it can be utilized in any type of data. The experiments for target detection applications on the hyperspectral images verify the effectiveness of the proposed method.
Structured Kernel Subspace Learning for Autonomous Robot Navigation.
Kim, Eunwoo; Choi, Sungjoon; Oh, Songhwai
2018-02-14
This paper considers two important problems for autonomous robot navigation in a dynamic environment, where the goal is to predict pedestrian motion and control a robot with the prediction for safe navigation. While there are several methods for predicting the motion of a pedestrian and controlling a robot to avoid incoming pedestrians, it is still difficult to safely navigate in a dynamic environment due to challenges, such as the varying quality and complexity of training data with unwanted noises. This paper addresses these challenges simultaneously by proposing a robust kernel subspace learning algorithm based on the recent advances in nuclear-norm and l 1 -norm minimization. We model the motion of a pedestrian and the robot controller using Gaussian processes. The proposed method efficiently approximates a kernel matrix used in Gaussian process regression by learning low-rank structured matrix (with symmetric positive semi-definiteness) to find an orthogonal basis, which eliminates the effects of erroneous and inconsistent data. Based on structured kernel subspace learning, we propose a robust motion model and motion controller for safe navigation in dynamic environments. We evaluate the proposed robust kernel learning in various tasks, including regression, motion prediction, and motion control problems, and demonstrate that the proposed learning-based systems are robust against outliers and outperform existing regression and navigation methods.
Yu, Yinan; Diamantaras, Konstantinos I; McKelvey, Tomas; Kung, Sun-Yuan
2018-02-01
In kernel-based classification models, given limited computational power and storage capacity, operations over the full kernel matrix becomes prohibitive. In this paper, we propose a new supervised learning framework using kernel models for sequential data processing. The framework is based on two components that both aim at enhancing the classification capability with a subset selection scheme. The first part is a subspace projection technique in the reproducing kernel Hilbert space using a CLAss-specific Subspace Kernel representation for kernel approximation. In the second part, we propose a novel structural risk minimization algorithm called the adaptive margin slack minimization to iteratively improve the classification accuracy by an adaptive data selection. We motivate each part separately, and then integrate them into learning frameworks for large scale data. We propose two such frameworks: the memory efficient sequential processing for sequential data processing and the parallelized sequential processing for distributed computing with sequential data acquisition. We test our methods on several benchmark data sets and compared with the state-of-the-art techniques to verify the validity of the proposed techniques.
Scalable Density-Based Subspace Clustering
Müller, Emmanuel; Assent, Ira; Günnemann, Stephan
2011-01-01
For knowledge discovery in high dimensional databases, subspace clustering detects clusters in arbitrary subspace projections. Scalability is a crucial issue, as the number of possible projections is exponential in the number of dimensions. We propose a scalable density-based subspace clustering...... method that steers mining to few selected subspace clusters. Our novel steering technique reduces subspace processing by identifying and clustering promising subspaces and their combinations directly. Thereby, it narrows down the search space while maintaining accuracy. Thorough experiments on real...... and synthetic databases show that steering is efficient and scalable, with high quality results. For future work, our steering paradigm for density-based subspace clustering opens research potential for speeding up other subspace clustering approaches as well....
Subspace Based Blind Sparse Channel Estimation
Hayashi, Kazunori; Matsushima, Hiroki; Sakai, Hideaki
2012-01-01
The paper proposes a subspace based blind sparse channel estimation method using 1–2 optimization by replacing the 2–norm minimization in the conventional subspace based method by the 1–norm minimization problem. Numerical results confirm that the proposed method can significantly improve...
Alim Samat
2016-03-01
Full Text Available In order to deal with scenarios where the training data, used to deduce a model, and the validation data have different statistical distributions, we study the problem of transformed subspace feature transfer for domain adaptation (DA in the context of hyperspectral image classification via a geodesic Gaussian flow kernel based support vector machine (GFKSVM. To show the superior performance of the proposed approach, conventional support vector machines (SVMs and state-of-the-art DA algorithms, including information-theoretical learning of discriminative cluster for domain adaptation (ITLDC, joint distribution adaptation (JDA, and joint transfer matching (JTM, are also considered. Additionally, unsupervised linear and nonlinear subspace feature transfer techniques including principal component analysis (PCA, randomized nonlinear principal component analysis (rPCA, factor analysis (FA and non-negative matrix factorization (NNMF are investigated and compared. Experiments on two real hyperspectral images show the cross-image classification performances of the GFKSVM, confirming its effectiveness and suitability when applied to hyperspectral images.
Arenas-Garcia, J.; Petersen, K.; Camps-Valls, G.
2013-01-01
correlation analysis (CCA), and orthonormalized PLS (OPLS), as well as their nonlinear extensions derived by means of the theory of reproducing kernel Hilbert spaces (RKHSs). We also review their connections to other methods for classification and statistical dependence estimation and introduce some recent...
Unsupervised spike sorting based on discriminative subspace learning.
Keshtkaran, Mohammad Reza; Yang, Zhi
2014-01-01
Spike sorting is a fundamental preprocessing step for many neuroscience studies which rely on the analysis of spike trains. In this paper, we present two unsupervised spike sorting algorithms based on discriminative subspace learning. The first algorithm simultaneously learns the discriminative feature subspace and performs clustering. It uses histogram of features in the most discriminative projection to detect the number of neurons. The second algorithm performs hierarchical divisive clustering that learns a discriminative 1-dimensional subspace for clustering in each level of the hierarchy until achieving almost unimodal distribution in the subspace. The algorithms are tested on synthetic and in-vivo data, and are compared against two widely used spike sorting methods. The comparative results demonstrate that our spike sorting methods can achieve substantially higher accuracy in lower dimensional feature space, and they are highly robust to noise. Moreover, they provide significantly better cluster separability in the learned subspace than in the subspace obtained by principal component analysis or wavelet transform.
Gradient-based adaptation of general gaussian kernels.
Glasmachers, Tobias; Igel, Christian
2005-10-01
Gradient-based optimizing of gaussian kernel functions is considered. The gradient for the adaptation of scaling and rotation of the input space is computed to achieve invariance against linear transformations. This is done by using the exponential map as a parameterization of the kernel parameter manifold. By restricting the optimization to a constant trace subspace, the kernel size can be controlled. This is, for example, useful to prevent overfitting when minimizing radius-margin generalization performance measures. The concepts are demonstrated by training hard margin support vector machines on toy data.
Efficient Online Subspace Learning With an Indefinite Kernel for Visual Tracking and Recognition
Liwicki, Stephan; Zafeiriou, Stefanos; Tzimiropoulos, Georgios; Pantic, Maja
2012-01-01
We propose an exact framework for online learning with a family of indefinite (not positive) kernels. As we study the case of nonpositive kernels, we first show how to extend kernel principal component analysis (KPCA) from a reproducing kernel Hilbert space to Krein space. We then formulate an
EVD Dualdating Based Online Subspace Learning
Bo Jin
2014-01-01
Full Text Available Conventional incremental PCA methods usually only discuss the situation of adding samples. In this paper, we consider two different cases: deleting samples and simultaneously adding and deleting samples. To avoid the NP-hard problem of downdating SVD without right singular vectors and specific position information, we choose to use EVD instead of SVD, which is used by most IPCA methods. First, we propose an EVD updating and downdating algorithm, called EVD dualdating, which permits simultaneous arbitrary adding and deleting operation, via transforming the EVD of the covariance matrix into a SVD updating problem plus an EVD of a small autocorrelation matrix. A comprehensive analysis is delivered to express the essence, expansibility, and computation complexity of EVD dualdating. A mathematical theorem proves that if the whole data matrix satisfies the low-rank-plus-shift structure, EVD dualdating is an optimal rank-k estimator under the sequential environment. A selection method based on eigenvalues is presented to determine the optimal rank k of the subspace. Then, we propose three incremental/decremental PCA methods: EVDD-IPCA, EVDD-DPCA, and EVDD-IDPCA, which are adaptive to the varying mean. Finally, plenty of comparative experiments demonstrate that EVDD-based methods outperform conventional incremental/decremental PCA methods in both efficiency and accuracy.
Roller Bearing Monitoring by New Subspace-Based Damage Indicator
G. Gautier
2015-01-01
Full Text Available A frequency-band subspace-based damage identification method for fault diagnosis in roller bearings is presented. Subspace-based damage indicators are obtained by filtering the vibration data in the frequency range where damage is likely to occur, that is, around the bearing characteristic frequencies. The proposed method is validated by considering simulated data of a damaged bearing. Also, an experimental case is considered which focuses on collecting the vibration data issued from a run-to-failure test. It is shown that the proposed method can detect bearing defects and, as such, it appears to be an efficient tool for diagnosis purpose.
Random matrix improved subspace clustering
Couillet, Romain
2017-03-06
This article introduces a spectral method for statistical subspace clustering. The method is built upon standard kernel spectral clustering techniques, however carefully tuned by theoretical understanding arising from random matrix findings. We show in particular that our method provides high clustering performance while standard kernel choices provably fail. An application to user grouping based on vector channel observations in the context of massive MIMO wireless communication networks is provided.
Subspace-based Inverse Uncertainty Quantification for Nuclear Data Assessment
Khuwaileh, B.A., E-mail: bakhuwai@ncsu.edu; Abdel-Khalik, H.S.
2015-01-15
Safety analysis and design optimization depend on the accurate prediction of various reactor attributes. Predictions can be enhanced by reducing the uncertainty associated with the attributes of interest. An inverse problem can be defined and solved to assess the sources of uncertainty, and experimental effort can be subsequently directed to further improve the uncertainty associated with these sources. In this work a subspace-based algorithm for inverse sensitivity/uncertainty quantification (IS/UQ) has been developed to enable analysts account for all sources of nuclear data uncertainties in support of target accuracy assessment-type analysis. An approximate analytical solution of the optimization problem is used to guide the search for the dominant uncertainty subspace. By limiting the search to a subspace, the degrees of freedom available for the optimization search are significantly reduced. A quarter PWR fuel assembly is modeled and the accuracy of the multiplication factor and the fission reaction rate are used as reactor attributes whose uncertainties are to be reduced. Numerical experiments are used to demonstrate the computational efficiency of the proposed algorithm. Our ongoing work is focusing on extending the proposed algorithm to account for various forms of feedback, e.g., thermal-hydraulics and depletion effects.
Zhang, Guoqing; Sun, Huaijiang; Xia, Guiyu; Sun, Quansen
2016-07-07
Sparse representation based classification (SRC) has been developed and shown great potential for real-world application. Based on SRC, Yang et al. [10] devised a SRC steered discriminative projection (SRC-DP) method. However, as a linear algorithm, SRC-DP cannot handle the data with highly nonlinear distribution. Kernel sparse representation-based classifier (KSRC) is a non-linear extension of SRC and can remedy the drawback of SRC. KSRC requires the use of a predetermined kernel function and selection of the kernel function and its parameters is difficult. Recently, multiple kernel learning for SRC (MKL-SRC) [22] has been proposed to learn a kernel from a set of base kernels. However, MKL-SRC only considers the within-class reconstruction residual while ignoring the between-class relationship, when learning the kernel weights. In this paper, we propose a novel multiple kernel sparse representation-based classifier (MKSRC), and then we use it as a criterion to design a multiple kernel sparse representation based orthogonal discriminative projection method (MK-SR-ODP). The proposed algorithm aims at learning a projection matrix and a corresponding kernel from the given base kernels such that in the low dimension subspace the between-class reconstruction residual is maximized and the within-class reconstruction residual is minimized. Furthermore, to achieve a minimum overall loss by performing recognition in the learned low-dimensional subspace, we introduce cost information into the dimensionality reduction method. The solutions for the proposed method can be efficiently found based on trace ratio optimization method [33]. Extensive experimental results demonstrate the superiority of the proposed algorithm when compared with the state-of-the-art methods.
Paramecium: An Extensible Object-Based Kernel
van Doorn, L.; Homburg, P.; Tanenbaum, A.S.
1995-01-01
In this paper we describe the design of an extensible kernel, called Paramecium. This kernel uses an object-based software architecture which together with instance naming, late binding and explicit overrides enables easy reconfiguration. Determining which components reside in the kernel protection
The influence of different PAST-based subspace trackers on DaPT parameter estimation
Lechtenberg, M.; Götze, J.
2012-09-01
In the context of parameter estimation, subspace-based methods like ESPRIT have become common. They require a subspace separation e.g. based on eigenvalue/-vector decomposition. In time-varying environments, this can be done by subspace trackers. One class of these is based on the PAST algorithm. Our non-linear parameter estimation algorithm DaPT builds on-top of the ESPRIT algorithm. Evaluation of the different variants of the PAST algorithm shows which variant of the PAST algorithm is worthwhile in the context of frequency estimation.
Experimental Comparison of Signal Subspace Based Noise Reduction Methods
Hansen, Peter Søren Kirk; Hansen, Per Christian; Hansen, Steffen Duus
1999-01-01
The signal subspace approach for non-parametric speech enhancement is considered. Several algorithms have been proposed in the literature but only partly analyzed. Here, the different algorithms are compared, and the emphasis is put onto the limiting factors and practical behavior of the estimators...
Senyue Zhang
2016-01-01
Full Text Available According to the characteristics that the kernel function of extreme learning machine (ELM and its performance have a strong correlation, a novel extreme learning machine based on a generalized triangle Hermitian kernel function was proposed in this paper. First, the generalized triangle Hermitian kernel function was constructed by using the product of triangular kernel and generalized Hermite Dirichlet kernel, and the proposed kernel function was proved as a valid kernel function of extreme learning machine. Then, the learning methodology of the extreme learning machine based on the proposed kernel function was presented. The biggest advantage of the proposed kernel is its kernel parameter values only chosen in the natural numbers, which thus can greatly shorten the computational time of parameter optimization and retain more of its sample data structure information. Experiments were performed on a number of binary classification, multiclassification, and regression datasets from the UCI benchmark repository. The experiment results demonstrated that the robustness and generalization performance of the proposed method are outperformed compared to other extreme learning machines with different kernels. Furthermore, the learning speed of proposed method is faster than support vector machine (SVM methods.
Random matrix improved subspace clustering
Couillet, Romain; Kammoun, Abla
2017-01-01
This article introduces a spectral method for statistical subspace clustering. The method is built upon standard kernel spectral clustering techniques, however carefully tuned by theoretical understanding arising from random matrix findings. We show
Approximate kernel competitive learning.
Wu, Jian-Sheng; Zheng, Wei-Shi; Lai, Jian-Huang
2015-03-01
Kernel competitive learning has been successfully used to achieve robust clustering. However, kernel competitive learning (KCL) is not scalable for large scale data processing, because (1) it has to calculate and store the full kernel matrix that is too large to be calculated and kept in the memory and (2) it cannot be computed in parallel. In this paper we develop a framework of approximate kernel competitive learning for processing large scale dataset. The proposed framework consists of two parts. First, it derives an approximate kernel competitive learning (AKCL), which learns kernel competitive learning in a subspace via sampling. We provide solid theoretical analysis on why the proposed approximation modelling would work for kernel competitive learning, and furthermore, we show that the computational complexity of AKCL is largely reduced. Second, we propose a pseudo-parallelled approximate kernel competitive learning (PAKCL) based on a set-based kernel competitive learning strategy, which overcomes the obstacle of using parallel programming in kernel competitive learning and significantly accelerates the approximate kernel competitive learning for large scale clustering. The empirical evaluation on publicly available datasets shows that the proposed AKCL and PAKCL can perform comparably as KCL, with a large reduction on computational cost. Also, the proposed methods achieve more effective clustering performance in terms of clustering precision against related approximate clustering approaches. Copyright © 2014 Elsevier Ltd. All rights reserved.
Subspace-Based Holistic Registration for Low-Resolution Facial Images
Boom BJ
2010-01-01
Full Text Available Subspace-based holistic registration is introduced as an alternative to landmark-based face registration, which has a poor performance on low-resolution images, as obtained in camera surveillance applications. The proposed registration method finds the alignment by maximizing the similarity score between a probe and a gallery image. We use a novel probabilistic framework for both user-independent as well as user-specific face registration. The similarity is calculated using the probability that the face image is correctly aligned in a face subspace, but additionally we take the probability into account that the face is misaligned based on the residual error in the dimensions perpendicular to the face subspace. We perform extensive experiments on the FRGCv2 database to evaluate the impact that the face registration methods have on face recognition. Subspace-based holistic registration on low-resolution images can improve face recognition in comparison with landmark-based registration on high-resolution images. The performance of the tested face recognition methods after subspace-based holistic registration on a low-resolution version of the FRGC database is similar to that after manual registration.
Subspace-based analysis of the ERT inverse problem
Ben Hadj Miled, Mohamed Khames; Miller, Eric L.
2004-05-01
In a previous work, we proposed a source-type formulation to the electrical resistance tomography (ERT) problem. Specifically, we showed that inhomogeneities in the medium can be viewed as secondary sources embedded in the homogeneous background medium and located at positions associated with variation in electrical conductivity. Assuming a piecewise constant conductivity distribution, the support of equivalent sources is equal to the boundary of the inhomogeneity. The estimation of the anomaly shape takes the form of an inverse source-type problem. In this paper, we explore the use of subspace methods to localize the secondary equivalent sources associated with discontinuities in the conductivity distribution. Our first alternative is the multiple signal classification (MUSIC) algorithm which is commonly used in the localization of multiple sources. The idea is to project a finite collection of plausible pole (or dipole) sources onto an estimated signal subspace and select those with largest correlations. In ERT, secondary sources are excited simultaneously but in different ways, i.e. with distinct amplitude patterns, depending on the locations and amplitudes of primary sources. If the number of receivers is "large enough", different source configurations can lead to a set of observation vectors that span the data subspace. However, since sources that are spatially close to each other have highly correlated signatures, seperation of such signals becomes very difficult in the presence of noise. To overcome this problem we consider iterative MUSIC algorithms like R-MUSIC and RAP-MUSIC. These recursive algorithms pose a computational burden as they require multiple large combinatorial searches. Results obtained with these algorithms using simulated data of different conductivity patterns are presented.
Estimation of direction of arrival of a moving target using subspace based approaches
Ghosh, Ripul; Das, Utpal; Akula, Aparna; Kumar, Satish; Sardana, H. K.
2016-05-01
In this work, array processing techniques based on subspace decomposition of signal have been evaluated for estimation of direction of arrival of moving targets using acoustic signatures. Three subspace based approaches - Incoherent Wideband Multiple Signal Classification (IWM), Least Square-Estimation of Signal Parameters via Rotation Invariance Techniques (LS-ESPRIT) and Total Least Square- ESPIRIT (TLS-ESPRIT) are considered. Their performance is compared with conventional time delay estimation (TDE) approaches such as Generalized Cross Correlation (GCC) and Average Square Difference Function (ASDF). Performance evaluation has been conducted on experimentally generated data consisting of acoustic signatures of four different types of civilian vehicles moving in defined geometrical trajectories. Mean absolute error and standard deviation of the DOA estimates w.r.t. ground truth are used as performance evaluation metrics. Lower statistical values of mean error confirm the superiority of subspace based approaches over TDE based techniques. Amongst the compared methods, LS-ESPRIT indicated better performance.
Subspace-Based Noise Reduction for Speech Signals via Diagonal and Triangular Matrix Decompositions
Hansen, Per Christian; Jensen, Søren Holdt
We survey the definitions and use of rank-revealing matrix decompositions in single-channel noise reduction algorithms for speech signals. Our algorithms are based on the rank-reduction paradigm and, in particular, signal subspace techniques. The focus is on practical working algorithms, using both...... diagonal (eigenvalue and singular value) decompositions and rank-revealing triangular decompositions (ULV, URV, VSV, ULLV and ULLIV). In addition we show how the subspace-based algorithms can be evaluated and compared by means of simple FIR filter interpretations. The algorithms are illustrated...... with working Matlab code and applications in speech processing....
Scuba: scalable kernel-based gene prioritization.
Zampieri, Guido; Tran, Dinh Van; Donini, Michele; Navarin, Nicolò; Aiolli, Fabio; Sperduti, Alessandro; Valle, Giorgio
2018-01-25
The uncovering of genes linked to human diseases is a pressing challenge in molecular biology and precision medicine. This task is often hindered by the large number of candidate genes and by the heterogeneity of the available information. Computational methods for the prioritization of candidate genes can help to cope with these problems. In particular, kernel-based methods are a powerful resource for the integration of heterogeneous biological knowledge, however, their practical implementation is often precluded by their limited scalability. We propose Scuba, a scalable kernel-based method for gene prioritization. It implements a novel multiple kernel learning approach, based on a semi-supervised perspective and on the optimization of the margin distribution. Scuba is optimized to cope with strongly unbalanced settings where known disease genes are few and large scale predictions are required. Importantly, it is able to efficiently deal both with a large amount of candidate genes and with an arbitrary number of data sources. As a direct consequence of scalability, Scuba integrates also a new efficient strategy to select optimal kernel parameters for each data source. We performed cross-validation experiments and simulated a realistic usage setting, showing that Scuba outperforms a wide range of state-of-the-art methods. Scuba achieves state-of-the-art performance and has enhanced scalability compared to existing kernel-based approaches for genomic data. This method can be useful to prioritize candidate genes, particularly when their number is large or when input data is highly heterogeneous. The code is freely available at https://github.com/gzampieri/Scuba .
Radjavi, Heydar
2003-01-01
This broad survey spans a wealth of studies on invariant subspaces, focusing on operators on separable Hilbert space. Largely self-contained, it requires only a working knowledge of measure theory, complex analysis, and elementary functional analysis. Subjects include normal operators, analytic functions of operators, shift operators, examples of invariant subspace lattices, compact operators, and the existence of invariant and hyperinvariant subspaces. Additional chapters cover certain results on von Neumann algebras, transitive operator algebras, algebras associated with invariant subspaces,
Subspace-Based Noise Reduction for Speech Signals via Diagonal and Triangular Matrix Decompositions
Hansen, Per Christian; Jensen, Søren Holdt
2007-01-01
We survey the definitions and use of rank-revealing matrix decompositions in single-channel noise reduction algorithms for speech signals. Our algorithms are based on the rank-reduction paradigm and, in particular, signal subspace techniques. The focus is on practical working algorithms, using both...... with working Matlab code and applications in speech processing....
Kernel-based tests for joint independence
Pfister, Niklas; Bühlmann, Peter; Schölkopf, Bernhard
2018-01-01
if the $d$ variables are jointly independent, as long as the kernel is characteristic. Based on an empirical estimate of dHSIC, we define three different non-parametric hypothesis tests: a permutation test, a bootstrap test and a test based on a Gamma approximation. We prove that the permutation test......We investigate the problem of testing whether $d$ random variables, which may or may not be continuous, are jointly (or mutually) independent. Our method builds on ideas of the two variable Hilbert-Schmidt independence criterion (HSIC) but allows for an arbitrary number of variables. We embed...... the $d$-dimensional joint distribution and the product of the marginals into a reproducing kernel Hilbert space and define the $d$-variable Hilbert-Schmidt independence criterion (dHSIC) as the squared distance between the embeddings. In the population case, the value of dHSIC is zero if and only...
Keshtkaran, Mohammad Reza; Yang, Zhi
2017-06-01
Spike sorting is a fundamental preprocessing step for many neuroscience studies which rely on the analysis of spike trains. Most of the feature extraction and dimensionality reduction techniques that have been used for spike sorting give a projection subspace which is not necessarily the most discriminative one. Therefore, the clusters which appear inherently separable in some discriminative subspace may overlap if projected using conventional feature extraction approaches leading to a poor sorting accuracy especially when the noise level is high. In this paper, we propose a noise-robust and unsupervised spike sorting algorithm based on learning discriminative spike features for clustering. The proposed algorithm uses discriminative subspace learning to extract low dimensional and most discriminative features from the spike waveforms and perform clustering with automatic detection of the number of the clusters. The core part of the algorithm involves iterative subspace selection using linear discriminant analysis and clustering using Gaussian mixture model with outlier detection. A statistical test in the discriminative subspace is proposed to automatically detect the number of the clusters. Comparative results on publicly available simulated and real in vivo datasets demonstrate that our algorithm achieves substantially improved cluster distinction leading to higher sorting accuracy and more reliable detection of clusters which are highly overlapping and not detectable using conventional feature extraction techniques such as principal component analysis or wavelets. By providing more accurate information about the activity of more number of individual neurons with high robustness to neural noise and outliers, the proposed unsupervised spike sorting algorithm facilitates more detailed and accurate analysis of single- and multi-unit activities in neuroscience and brain machine interface studies.
Keshtkaran, Mohammad Reza; Yang, Zhi
2017-06-01
Objective. Spike sorting is a fundamental preprocessing step for many neuroscience studies which rely on the analysis of spike trains. Most of the feature extraction and dimensionality reduction techniques that have been used for spike sorting give a projection subspace which is not necessarily the most discriminative one. Therefore, the clusters which appear inherently separable in some discriminative subspace may overlap if projected using conventional feature extraction approaches leading to a poor sorting accuracy especially when the noise level is high. In this paper, we propose a noise-robust and unsupervised spike sorting algorithm based on learning discriminative spike features for clustering. Approach. The proposed algorithm uses discriminative subspace learning to extract low dimensional and most discriminative features from the spike waveforms and perform clustering with automatic detection of the number of the clusters. The core part of the algorithm involves iterative subspace selection using linear discriminant analysis and clustering using Gaussian mixture model with outlier detection. A statistical test in the discriminative subspace is proposed to automatically detect the number of the clusters. Main results. Comparative results on publicly available simulated and real in vivo datasets demonstrate that our algorithm achieves substantially improved cluster distinction leading to higher sorting accuracy and more reliable detection of clusters which are highly overlapping and not detectable using conventional feature extraction techniques such as principal component analysis or wavelets. Significance. By providing more accurate information about the activity of more number of individual neurons with high robustness to neural noise and outliers, the proposed unsupervised spike sorting algorithm facilitates more detailed and accurate analysis of single- and multi-unit activities in neuroscience and brain machine interface studies.
Chen, Xudong
2010-01-01
This paper proposes a version of the subspace-based optimization method to solve the inverse scattering problem with an inhomogeneous background medium where the known inhomogeneities are bounded in a finite domain. Although the background Green's function at each discrete point in the computational domain is not directly available in an inhomogeneous background scenario, the paper uses the finite element method to simultaneously obtain the Green's function at all discrete points. The essence of the subspace-based optimization method is that part of the contrast source is determined from the spectrum analysis without using any optimization, whereas the orthogonally complementary part is determined by solving a lower dimension optimization problem. This feature significantly speeds up the convergence of the algorithm and at the same time makes it robust against noise. Numerical simulations illustrate the efficacy of the proposed algorithm. The algorithm presented in this paper finds wide applications in nondestructive evaluation, such as through-wall imaging
Kernel based eigenvalue-decomposition methods for analysing ham
Christiansen, Asger Nyman; Nielsen, Allan Aasbjerg; Møller, Flemming
2010-01-01
methods, such as PCA, MAF or MNF. We therefore investigated the applicability of kernel based versions of these transformation. This meant implementing the kernel based methods and developing new theory, since kernel based MAF and MNF is not described in the literature yet. The traditional methods only...... have two factors that are useful for segmentation and none of them can be used to segment the two types of meat. The kernel based methods have a lot of useful factors and they are able to capture the subtle differences in the images. This is illustrated in Figure 1. You can see a comparison of the most...... useful factor of PCA and kernel based PCA respectively in Figure 2. The factor of the kernel based PCA turned out to be able to segment the two types of meat and in general that factor is much more distinct, compared to the traditional factor. After the orthogonal transformation a simple thresholding...
Liesen, J.; Strakoš, Zdeněk
2008-01-01
Roč. 50, č. 3 (2008), s. 485-503 ISSN 0036-1445 R&D Projects: GA AV ČR 1ET400300415; GA AV ČR IAA100300802 Institutional research plan: CEZ:AV0Z10300504 Keywords : Krylov subspace methods * orthogonal bases * short reccurences * conjugate gradient -like methods Subject RIV: IN - Informatics, Computer Science Impact factor: 2.739, year: 2008
Hankel Matrix Correlation Function-Based Subspace Identification Method for UAV Servo System
Minghong She
2018-01-01
Full Text Available For the identification problem of closed-loop subspace model, we propose a zero space projection method based on the estimation of correlation function to fill the block Hankel matrix of identification model by combining the linear algebra with geometry. By using the same projection of related data in time offset set and LQ decomposition, the multiplication operation of projection is achieved and dynamics estimation of the unknown equipment system model is obtained. Consequently, we have solved the problem of biased estimation caused when the open-loop subspace identification algorithm is applied to the closed-loop identification. A simulation example is given to show the effectiveness of the proposed approach. In final, the practicability of the identification algorithm is verified by hardware test of UAV servo system in real environment.
Subspace based adaptive denoising of surface EMG from neurological injury patients
Liu, Jie; Ying, Dongwen; Zev Rymer, William; Zhou, Ping
2014-10-01
Objective: After neurological injuries such as spinal cord injury, voluntary surface electromyogram (EMG) signals recorded from affected muscles are often corrupted by interferences, such as spurious involuntary spikes and background noises produced by physiological and extrinsic/accidental origins, imposing difficulties for signal processing. Conventional methods did not well address the problem caused by interferences. It is difficult to mitigate such interferences using conventional methods. The aim of this study was to develop a subspace-based denoising method to suppress involuntary background spikes contaminating voluntary surface EMG recordings. Approach: The Karhunen-Loeve transform was utilized to decompose a noisy signal into a signal subspace and a noise subspace. An optimal estimate of EMG signal is derived from the signal subspace and the noise power. Specifically, this estimator is capable of making a tradeoff between interference reduction and signal distortion. Since the estimator partially relies on the estimate of noise power, an adaptive method was presented to sequentially track the variation of interference power. The proposed method was evaluated using both semi-synthetic and real surface EMG signals. Main results: The experiments confirmed that the proposed method can effectively suppress interferences while keep the distortion of voluntary EMG signal in a low level. The proposed method can greatly facilitate further signal processing, such as onset detection of voluntary muscle activity. Significance: The proposed method can provide a powerful tool for suppressing background spikes and noise contaminating voluntary surface EMG signals of paretic muscles after neurological injuries, which is of great importance for their multi-purpose applications.
N-screen aware multicriteria hybrid recommender system using weight based subspace clustering.
Ullah, Farman; Sarwar, Ghulam; Lee, Sungchang
2014-01-01
This paper presents a recommender system for N-screen services in which users have multiple devices with different capabilities. In N-screen services, a user can use various devices in different locations and time and can change a device while the service is running. N-screen aware recommendation seeks to improve the user experience with recommended content by considering the user N-screen device attributes such as screen resolution, media codec, remaining battery time, and access network and the user temporal usage pattern information that are not considered in existing recommender systems. For N-screen aware recommendation support, this work introduces a user device profile collaboration agent, manager, and N-screen control server to acquire and manage the user N-screen devices profile. Furthermore, a multicriteria hybrid framework is suggested that incorporates the N-screen devices information with user preferences and demographics. In addition, we propose an individual feature and subspace weight based clustering (IFSWC) to assign different weights to each subspace and each feature within a subspace in the hybrid framework. The proposed system improves the accuracy, precision, scalability, sparsity, and cold start issues. The simulation results demonstrate the effectiveness and prove the aforementioned statements.
A new subspace based approach to iterative learning control
Nijsse, G.; Verhaegen, M.; Doelman, N.J.
2001-01-01
This paper1 presents an iterative learning control (ILC) procedure based on an inverse model of the plant under control. Our first contribution is that we formulate the inversion procedure as a Kalman smoothing problem: based on a compact state space model of a possibly non-minimum phase system,
A Comfort-Aware Energy Efficient HVAC System Based on the Subspace Identification Method
O. Tsakiridis
2016-01-01
Full Text Available A proactive heating method is presented aiming at reducing the energy consumption in a HVAC system while maintaining the thermal comfort of the occupants. The proposed technique fuses time predictions for the zones’ temperatures, based on a deterministic subspace identification method, and zones’ occupancy predictions, based on a mobility model, in a decision scheme that is capable of regulating the balance between the total energy consumed and the total discomfort cost. Simulation results for various occupation-mobility models demonstrate the efficiency of the proposed technique.
Subspace-based interference removal methods for a multichannel biomagnetic sensor array
Sekihara, Kensuke; Nagarajan, Srikantan S.
2017-10-01
Objective. In biomagnetic signal processing, the theory of the signal subspace has been applied to removing interfering magnetic fields, and a representative algorithm is the signal space projection algorithm, in which the signal/interference subspace is defined in the spatial domain as the span of signal/interference-source lead field vectors. This paper extends the notion of this conventional (spatial domain) signal subspace by introducing a new definition of signal subspace in the time domain. Approach. It defines the time-domain signal subspace as the span of row vectors that contain the source time course values. This definition leads to symmetric relationships between the time-domain and the conventional (spatial-domain) signal subspaces. As a review, this article shows that the notion of the time-domain signal subspace provides useful insights over existing interference removal methods from a unified perspective. Main results and significance. Using the time-domain signal subspace, it is possible to interpret a number of interference removal methods as the time domain signal space projection. Such methods include adaptive noise canceling, sensor noise suppression, the common temporal subspace projection, the spatio-temporal signal space separation, and the recently-proposed dual signal subspace projection. Our analysis using the notion of the time domain signal space projection reveals implicit assumptions these methods rely on, and shows that the difference between these methods results only from the manner of deriving the interference subspace. Numerical examples that illustrate the results of our arguments are provided.
Consistency analysis of subspace identification methods based on a linear regression approach
Knudsen, Torben
2001-01-01
In the literature results can be found which claim consistency for the subspace method under certain quite weak assumptions. Unfortunately, a new result gives a counter example showing inconsistency under these assumptions and then gives new more strict sufficient assumptions which however does n...... not include important model structures as e.g. Box-Jenkins. Based on a simple least squares approach this paper shows the possible inconsistency under the weak assumptions and develops only slightly stricter assumptions sufficient for consistency and which includes any model structure...
An Improved EMD-Based Dissimilarity Metric for Unsupervised Linear Subspace Learning
Xiangchun Yu
2018-01-01
Full Text Available We investigate a novel way of robust face image feature extraction by adopting the methods based on Unsupervised Linear Subspace Learning to extract a small number of good features. Firstly, the face image is divided into blocks with the specified size, and then we propose and extract pooled Histogram of Oriented Gradient (pHOG over each block. Secondly, an improved Earth Mover’s Distance (EMD metric is adopted to measure the dissimilarity between blocks of one face image and the corresponding blocks from the rest of face images. Thirdly, considering the limitations of the original Locality Preserving Projections (LPP, we proposed the Block Structure LPP (BSLPP, which effectively preserves the structural information of face images. Finally, an adjacency graph is constructed and a small number of good features of a face image are obtained by methods based on Unsupervised Linear Subspace Learning. A series of experiments have been conducted on several well-known face databases to evaluate the effectiveness of the proposed algorithm. In addition, we construct the noise, geometric distortion, slight translation, slight rotation AR, and Extended Yale B face databases, and we verify the robustness of the proposed algorithm when faced with a certain degree of these disturbances.
Chen, Peng; Yang, Yixin; Wang, Yong; Ma, Yuanliang
2018-05-08
When sensor position errors exist, the performance of recently proposed interference-plus-noise covariance matrix (INCM)-based adaptive beamformers may be severely degraded. In this paper, we propose a weighted subspace fitting-based INCM reconstruction algorithm to overcome sensor displacement for linear arrays. By estimating the rough signal directions, we construct a novel possible mismatched steering vector (SV) set. We analyze the proximity of the signal subspace from the sample covariance matrix (SCM) and the space spanned by the possible mismatched SV set. After solving an iterative optimization problem, we reconstruct the INCM using the estimated sensor position errors. Then we estimate the SV of the desired signal by solving an optimization problem with the reconstructed INCM. The main advantage of the proposed algorithm is its robustness against SV mismatches dominated by unknown sensor position errors. Numerical examples show that even if the position errors are up to half of the assumed sensor spacing, the output signal-to-interference-plus-noise ratio is only reduced by 4 dB. Beam patterns plotted using experiment data show that the interference suppression capability of the proposed beamformer outperforms other tested beamformers.
Multiple kernel boosting framework based on information measure for classification
Qi, Chengming; Wang, Yuping; Tian, Wenjie; Wang, Qun
2016-01-01
The performance of kernel-based method, such as support vector machine (SVM), is greatly affected by the choice of kernel function. Multiple kernel learning (MKL) is a promising family of machine learning algorithms and has attracted many attentions in recent years. MKL combines multiple sub-kernels to seek better results compared to single kernel learning. In order to improve the efficiency of SVM and MKL, in this paper, the Kullback–Leibler kernel function is derived to develop SVM. The proposed method employs an improved ensemble learning framework, named KLMKB, which applies Adaboost to learning multiple kernel-based classifier. In the experiment for hyperspectral remote sensing image classification, we employ feature selected through Optional Index Factor (OIF) to classify the satellite image. We extensively examine the performance of our approach in comparison to some relevant and state-of-the-art algorithms on a number of benchmark classification data sets and hyperspectral remote sensing image data set. Experimental results show that our method has a stable behavior and a noticeable accuracy for different data set.
Rank-defective millimeter-wave channel estimation based on subspace-compressive sensing
Majid Shakhsi Dastgahian
2016-11-01
Full Text Available Millimeter-wave communication (mmWC is considered as one of the pioneer candidates for 5G indoor and outdoor systems in E-band. To subdue the channel propagation characteristics in this band, high dimensional antenna arrays need to be deployed at both the base station (BS and mobile sets (MS. Unlike the conventional MIMO systems, Millimeter-wave (mmW systems lay away to employ the power predatory equipment such as ADC or RF chain in each branch of MIMO system because of hardware constraints. Such systems leverage to the hybrid precoding (combining architecture for downlink deployment. Because there is a large array at the transceiver, it is impossible to estimate the channel by conventional methods. This paper develops a new algorithm to estimate the mmW channel by exploiting the sparse nature of the channel. The main contribution is the representation of a sparse channel model and the exploitation of a modified approach based on Multiple Measurement Vector (MMV greedy sparse framework and subspace method of Multiple Signal Classification (MUSIC which work together to recover the indices of non-zero elements of an unknown channel matrix when the rank of the channel matrix is defected. In practical rank-defective channels, MUSIC fails, and we need to propose new extended MUSIC approaches based on subspace enhancement to compensate the limitation of MUSIC. Simulation results indicate that our proposed extended MUSIC algorithms will have proper performances and moderate computational speeds, and that they are even able to work in channels with an unknown sparsity level.
Kernel-based noise filtering of neutron detector signals
Park, Moon Ghu; Shin, Ho Cheol; Lee, Eun Ki
2007-01-01
This paper describes recently developed techniques for effective filtering of neutron detector signal noise. In this paper, three kinds of noise filters are proposed and their performance is demonstrated for the estimation of reactivity. The tested filters are based on the unilateral kernel filter, unilateral kernel filter with adaptive bandwidth and bilateral filter to show their effectiveness in edge preservation. Filtering performance is compared with conventional low-pass and wavelet filters. The bilateral filter shows a remarkable improvement compared with unilateral kernel and wavelet filters. The effectiveness and simplicity of the unilateral kernel filter with adaptive bandwidth is also demonstrated by applying it to the reactivity measurement performed during reactor start-up physics tests
Acoustic Source Localization via Subspace Based Method Using Small Aperture MEMS Arrays
Xin Zhang
2014-01-01
Full Text Available Small aperture microphone arrays provide many advantages for portable devices and hearing aid equipment. In this paper, a subspace based localization method is proposed for acoustic source using small aperture arrays. The effects of array aperture on localization are analyzed by using array response (array manifold. Besides array aperture, the frequency of acoustic source and the variance of signal power are simulated to demonstrate how to optimize localization performance, which is carried out by introducing frequency error with the proposed method. The proposed method for 5 mm array aperture is validated by simulations and experiments with MEMS microphone arrays. Different types of acoustic sources can be localized with the highest precision of 6 degrees even in the presence of wind noise and other noises. Furthermore, the proposed method reduces the computational complexity compared with other methods.
Local coding based matching kernel method for image classification.
Yan Song
Full Text Available This paper mainly focuses on how to effectively and efficiently measure visual similarity for local feature based representation. Among existing methods, metrics based on Bag of Visual Word (BoV techniques are efficient and conceptually simple, at the expense of effectiveness. By contrast, kernel based metrics are more effective, but at the cost of greater computational complexity and increased storage requirements. We show that a unified visual matching framework can be developed to encompass both BoV and kernel based metrics, in which local kernel plays an important role between feature pairs or between features and their reconstruction. Generally, local kernels are defined using Euclidean distance or its derivatives, based either explicitly or implicitly on an assumption of Gaussian noise. However, local features such as SIFT and HoG often follow a heavy-tailed distribution which tends to undermine the motivation behind Euclidean metrics. Motivated by recent advances in feature coding techniques, a novel efficient local coding based matching kernel (LCMK method is proposed. This exploits the manifold structures in Hilbert space derived from local kernels. The proposed method combines advantages of both BoV and kernel based metrics, and achieves a linear computational complexity. This enables efficient and scalable visual matching to be performed on large scale image sets. To evaluate the effectiveness of the proposed LCMK method, we conduct extensive experiments with widely used benchmark datasets, including 15-Scenes, Caltech101/256, PASCAL VOC 2007 and 2011 datasets. Experimental results confirm the effectiveness of the relatively efficient LCMK method.
Ren, W. X.; Lin, Y. Q.; Fang, S. E.
2011-11-01
One of the key issues in vibration-based structural health monitoring is to extract the damage-sensitive but environment-insensitive features from sampled dynamic response measurements and to carry out the statistical analysis of these features for structural damage detection. A new damage feature is proposed in this paper by using the system matrices of the forward innovation model based on the covariance-driven stochastic subspace identification of a vibrating system. To overcome the variations of the system matrices, a non-singularity transposition matrix is introduced so that the system matrices are normalized to their standard forms. For reducing the effects of modeling errors, noise and environmental variations on measured structural responses, a statistical pattern recognition paradigm is incorporated into the proposed method. The Mahalanobis and Euclidean distance decision functions of the damage feature vector are adopted by defining a statistics-based damage index. The proposed structural damage detection method is verified against one numerical signal and two numerical beams. It is demonstrated that the proposed statistics-based damage index is sensitive to damage and shows some robustness to the noise and false estimation of the system ranks. The method is capable of locating damage of the beam structures under different types of excitations. The robustness of the proposed damage detection method to the variations in environmental temperature is further validated in a companion paper by a reinforced concrete beam tested in the laboratory and a full-scale arch bridge tested in the field.
Predictor-Year Subspace Clustering Based Ensemble Prediction of Indian Summer Monsoon
Moumita Saha
2016-01-01
Full Text Available Forecasting the Indian summer monsoon is a challenging task due to its complex and nonlinear behavior. A large number of global climatic variables with varying interaction patterns over years influence monsoon. Various statistical and neural prediction models have been proposed for forecasting monsoon, but many of them fail to capture variability over years. The skill of predictor variables of monsoon also evolves over time. In this article, we propose a joint-clustering of monsoon years and predictors for understanding and predicting the monsoon. This is achieved by subspace clustering algorithm. It groups the years based on prevailing global climatic condition using statistical clustering technique and subsequently for each such group it identifies significant climatic predictor variables which assist in better prediction. Prediction model is designed to frame individual cluster using random forest of regression tree. Prediction of aggregate and regional monsoon is attempted. Mean absolute error of 5.2% is obtained for forecasting aggregate Indian summer monsoon. Errors in predicting the regional monsoons are also comparable in comparison to the high variation of regional precipitation. Proposed joint-clustering based ensemble model is observed to be superior to existing monsoon prediction models and it also surpasses general nonclustering based prediction models.
Fault Diagnosis for Hydraulic Servo System Using Compressed Random Subspace Based ReliefF
Yu Ding
2018-01-01
Full Text Available Playing an important role in electromechanical systems, hydraulic servo system is crucial to mechanical systems like engineering machinery, metallurgical machinery, ships, and other equipment. Fault diagnosis based on monitoring and sensory signals plays an important role in avoiding catastrophic accidents and enormous economic losses. This study presents a fault diagnosis scheme for hydraulic servo system using compressed random subspace based ReliefF (CRSR method. From the point of view of feature selection, the scheme utilizes CRSR method to determine the most stable feature combination that contains the most adequate information simultaneously. Based on the feature selection structure of ReliefF, CRSR employs feature integration rules in the compressed domain. Meanwhile, CRSR substitutes information entropy and fuzzy membership for traditional distance measurement index. The proposed CRSR method is able to enhance the robustness of the feature information against interference while selecting the feature combination with balanced information expressing ability. To demonstrate the effectiveness of the proposed CRSR method, a hydraulic servo system joint simulation model is constructed by HyPneu and Simulink, and three fault modes are injected to generate the validation data.
Reduced multiple empirical kernel learning machine.
Wang, Zhe; Lu, MingZhe; Gao, Daqi
2015-02-01
Multiple kernel learning (MKL) is demonstrated to be flexible and effective in depicting heterogeneous data sources since MKL can introduce multiple kernels rather than a single fixed kernel into applications. However, MKL would get a high time and space complexity in contrast to single kernel learning, which is not expected in real-world applications. Meanwhile, it is known that the kernel mapping ways of MKL generally have two forms including implicit kernel mapping and empirical kernel mapping (EKM), where the latter is less attracted. In this paper, we focus on the MKL with the EKM, and propose a reduced multiple empirical kernel learning machine named RMEKLM for short. To the best of our knowledge, it is the first to reduce both time and space complexity of the MKL with EKM. Different from the existing MKL, the proposed RMEKLM adopts the Gauss Elimination technique to extract a set of feature vectors, which is validated that doing so does not lose much information of the original feature space. Then RMEKLM adopts the extracted feature vectors to span a reduced orthonormal subspace of the feature space, which is visualized in terms of the geometry structure. It can be demonstrated that the spanned subspace is isomorphic to the original feature space, which means that the dot product of two vectors in the original feature space is equal to that of the two corresponding vectors in the generated orthonormal subspace. More importantly, the proposed RMEKLM brings a simpler computation and meanwhile needs a less storage space, especially in the processing of testing. Finally, the experimental results show that RMEKLM owns a much efficient and effective performance in terms of both complexity and classification. The contributions of this paper can be given as follows: (1) by mapping the input space into an orthonormal subspace, the geometry of the generated subspace is visualized; (2) this paper first reduces both the time and space complexity of the EKM-based MKL; (3
Development of a Burnup Module DECBURN Based on the Krylov Subspace Method
Cho, J. Y.; Kim, K. S.; Shim, H. J.; Song, J. S
2008-05-15
This report is to develop a burnup module DECBURN that is essential for the reactor analysis and the assembly homogenization codes to trace the fuel composition change during the core burnup. The developed burnup module solves the burnup equation by the matrix exponential method based on the Krylov Subspace method. The final solution of the matrix exponential is obtained by the matrix scaling and squaring method. To develop DECBURN module, this report includes the followings as: (1) Krylov Subspace Method for Burnup Equation, (2) Manufacturing of the DECBURN module, (3) Library Structure Setup and Library Manufacturing, (4) Examination of the DECBURN module, (5) Implementation to the DeCART code and Verification. DECBURN library includes the decay constants, one-group cross section and the fission yields. Examination of the DECBURN module is performed by manufacturing a driver program, and the results of the DECBURN module is compared with those of the ORIGEN program. Also, the implemented DECBURN module to the DeCART code is applied to the LWR depletion benchmark and a OPR-1000 pin cell problem, and the solutions are compared with the HELIOS code to verify the computational soundness and accuracy. In this process, the criticality calculation method and the predictor-corrector scheme are introduced to the DeCART code for a function of the homogenization code. The examination by a driver program shows that the DECBURN module produces exactly the same solution with the ORIGEN program. DeCART code that equips the DECBURN module produces a compatible solution to the other codes for the LWR depletion benchmark. Also the multiplication factors of the DeCART code for the OPR-1000 pin cell problem agree to the HELIOS code within 100 pcm over the whole burnup steps. The multiplication factors with the criticality calculation are also compatible with the HELIOS code. These results mean that the developed DECBURN module works soundly and produces an accurate solution
Tatu, Aditya Jayant
This thesis deals with two unrelated issues, restricting curve evolution to subspaces and computing image patches in the equivalence class of Histogram of Gradient orientation based features using nonlinear projection methods. Curve evolution is a well known method used in various applications like...... tracking interfaces, active contour based segmentation methods and others. It can also be used to study shape spaces, as deforming a shape can be thought of as evolving its boundary curve. During curve evolution a curve traces out a path in the infinite dimensional space of curves. Due to application...... specific requirements like shape priors or a given data model, and due to limitations of the computer, the computed curve evolution forms a path in some finite dimensional subspace of the space of curves. We give methods to restrict the curve evolution to a finite dimensional linear or implicitly defined...
Similarity measurement method of high-dimensional data based on normalized net lattice subspace
Li Wenfa; Wang Gongming; Li Ke; Huang Su
2017-01-01
The performance of conventional similarity measurement methods is affected seriously by the curse of dimensionality of high-dimensional data.The reason is that data difference between sparse and noisy dimensionalities occupies a large proportion of the similarity, leading to the dissimilarities between any results.A similarity measurement method of high-dimensional data based on normalized net lattice subspace is proposed.The data range of each dimension is divided into several intervals, and the components in different dimensions are mapped onto the corresponding interval.Only the component in the same or adjacent interval is used to calculate the similarity.To validate this meth-od, three data types are used, and seven common similarity measurement methods are compared. The experimental result indicates that the relative difference of the method is increasing with the di-mensionality and is approximately two or three orders of magnitude higher than the conventional method.In addition, the similarity range of this method in different dimensions is [0, 1], which is fit for similarity analysis after dimensionality reduction.
A unified classifier for robust face recognition based on combining multiple subspace algorithms
Ijaz Bajwa, Usama; Ahmad Taj, Imtiaz; Waqas Anwar, Muhammad
2012-10-01
Face recognition being the fastest growing biometric technology has expanded manifold in the last few years. Various new algorithms and commercial systems have been proposed and developed. However, none of the proposed or developed algorithm is a complete solution because it may work very well on one set of images with say illumination changes but may not work properly on another set of image variations like expression variations. This study is motivated by the fact that any single classifier cannot claim to show generally better performance against all facial image variations. To overcome this shortcoming and achieve generality, combining several classifiers using various strategies has been studied extensively also incorporating the question of suitability of any classifier for this task. The study is based on the outcome of a comprehensive comparative analysis conducted on a combination of six subspace extraction algorithms and four distance metrics on three facial databases. The analysis leads to the selection of the most suitable classifiers which performs better on one task or the other. These classifiers are then combined together onto an ensemble classifier by two different strategies of weighted sum and re-ranking. The results of the ensemble classifier show that these strategies can be effectively used to construct a single classifier that can successfully handle varying facial image conditions of illumination, aging and facial expressions.
Chen Shi
2014-01-01
Full Text Available Subsynchronous oscillation (SSO usually caused by series compensation, power system stabilizer (PSS, high voltage direct current transmission (HVDC and other power electronic equipment, which will affect the safe operation of generator shafting even the system. It is very important to identify the modal parameters of SSO to take effective control strategies as well. Since the identification accuracy of traditional methods are not high enough, the stochastic subspace identification (SSI method is proposed to improve the identification accuracy of subsynchronous oscillation modal. The stochastic subspace identification method was compared with the other two methods on subsynchronous oscillation IEEE benchmark model and Xiang-Shang HVDC system model, the simulation results show that the stochastic subspace identification method has the advantages of high identification precision, high operation efficiency and strong ability of anti-noise.
2016-09-01
We consider the problem of subspace clustering: given points that lie on or near the union of many low-dimensional linear subspaces, recover the subspaces. To this end, one first identifies sets of points close to the same subspace and uses the sets ...
Banerjee, Amartya S; Lin, Lin; Suryanarayana, Phanish; Yang, Chao; Pask, John E
2018-06-12
We describe a novel iterative strategy for Kohn-Sham density functional theory calculations aimed at large systems (>1,000 electrons), applicable to metals and insulators alike. In lieu of explicit diagonalization of the Kohn-Sham Hamiltonian on every self-consistent field (SCF) iteration, we employ a two-level Chebyshev polynomial filter based complementary subspace strategy to (1) compute a set of vectors that span the occupied subspace of the Hamiltonian; (2) reduce subspace diagonalization to just partially occupied states; and (3) obtain those states in an efficient, scalable manner via an inner Chebyshev filter iteration. By reducing the necessary computation to just partially occupied states and obtaining these through an inner Chebyshev iteration, our approach reduces the cost of large metallic calculations significantly, while eliminating subspace diagonalization for insulating systems altogether. We describe the implementation of the method within the framework of the discontinuous Galerkin (DG) electronic structure method and show that this results in a computational scheme that can effectively tackle bulk and nano systems containing tens of thousands of electrons, with chemical accuracy, within a few minutes or less of wall clock time per SCF iteration on large-scale computing platforms. We anticipate that our method will be instrumental in pushing the envelope of large-scale ab initio molecular dynamics. As a demonstration of this, we simulate a bulk silicon system containing 8,000 atoms at finite temperature, and obtain an average SCF step wall time of 51 s on 34,560 processors; thus allowing us to carry out 1.0 ps of ab initio molecular dynamics in approximately 28 h (of wall time).
Synergistic Instance-Level Subspace Alignment for Fine-Grained Sketch-Based Image Retrieval.
Li, Ke; Pang, Kaiyue; Song, Yi-Zhe; Hospedales, Timothy M; Xiang, Tao; Zhang, Honggang
2017-08-25
We study the problem of fine-grained sketch-based image retrieval. By performing instance-level (rather than category-level) retrieval, it embodies a timely and practical application, particularly with the ubiquitous availability of touchscreens. Three factors contribute to the challenging nature of the problem: (i) free-hand sketches are inherently abstract and iconic, making visual comparisons with photos difficult, (ii) sketches and photos are in two different visual domains, i.e. black and white lines vs. color pixels, and (iii) fine-grained distinctions are especially challenging when executed across domain and abstraction-level. To address these challenges, we propose to bridge the image-sketch gap both at the high-level via parts and attributes, as well as at the low-level, via introducing a new domain alignment method. More specifically, (i) we contribute a dataset with 304 photos and 912 sketches, where each sketch and image is annotated with its semantic parts and associated part-level attributes. With the help of this dataset, we investigate (ii) how strongly-supervised deformable part-based models can be learned that subsequently enable automatic detection of part-level attributes, and provide pose-aligned sketch-image comparisons. To reduce the sketch-image gap when comparing low-level features, we also (iii) propose a novel method for instance-level domain-alignment, that exploits both subspace and instance-level cues to better align the domains. Finally (iv) these are combined in a matching framework integrating aligned low-level features, mid-level geometric structure and high-level semantic attributes. Extensive experiments conducted on our new dataset demonstrate effectiveness of the proposed method.
Xu, Y; Li, N
2014-09-01
Biological species have produced many simple but efficient rules in their complex and critical survival activities such as hunting and mating. A common feature observed in several biological motion strategies is that the predator only moves along paths in a carefully selected or iteratively refined subspace (or manifold), which might be able to explain why these motion strategies are effective. In this paper, a unified linear algebraic formulation representing such a predator-prey relationship is developed to simplify the construction and refinement process of the subspace (or manifold). Specifically, the following three motion strategies are studied and modified: motion camouflage, constant absolute target direction and local pursuit. The framework constructed based on this varying subspace concept could significantly reduce the computational cost in solving a class of nonlinear constrained optimal trajectory planning problems, particularly for the case with severe constraints. Two non-trivial examples, a ground robot and a hypersonic aircraft trajectory optimization problem, are used to show the capabilities of the algorithms in this new computational framework.
Xu, Y; Li, N
2014-01-01
Biological species have produced many simple but efficient rules in their complex and critical survival activities such as hunting and mating. A common feature observed in several biological motion strategies is that the predator only moves along paths in a carefully selected or iteratively refined subspace (or manifold), which might be able to explain why these motion strategies are effective. In this paper, a unified linear algebraic formulation representing such a predator–prey relationship is developed to simplify the construction and refinement process of the subspace (or manifold). Specifically, the following three motion strategies are studied and modified: motion camouflage, constant absolute target direction and local pursuit. The framework constructed based on this varying subspace concept could significantly reduce the computational cost in solving a class of nonlinear constrained optimal trajectory planning problems, particularly for the case with severe constraints. Two non-trivial examples, a ground robot and a hypersonic aircraft trajectory optimization problem, are used to show the capabilities of the algorithms in this new computational framework. (paper)
Imaging of heart acoustic based on the sub-space methods using a microphone array.
Moghaddasi, Hanie; Almasganj, Farshad; Zoroufian, Arezoo
2017-07-01
Heart disease is one of the leading causes of death around the world. Phonocardiogram (PCG) is an important bio-signal which represents the acoustic activity of heart, typically without any spatiotemporal information of the involved acoustic sources. The aim of this study is to analyze the PCG by employing a microphone array by which the heart internal sound sources could be localized, too. In this paper, it is intended to propose a modality by which the locations of the active sources in the heart could also be investigated, during a cardiac cycle. In this way, a microphone array with six microphones is employed as the recording set up to be put on the human chest. In the following, the Group Delay MUSIC algorithm which is a sub-space based localization method is used to estimate the location of the heart sources in different phases of the PCG. We achieved to 0.14cm mean error for the sources of first heart sound (S 1 ) simulator and 0.21cm mean error for the sources of second heart sound (S 2 ) simulator with Group Delay MUSIC algorithm. The acoustical diagrams created for human subjects show distinct patterns in various phases of the cardiac cycles such as the first and second heart sounds. Moreover, the evaluated source locations for the heart valves are matched with the ones that are obtained via the 4-dimensional (4D) echocardiography applied, to a real human case. Imaging of heart acoustic map presents a new outlook to indicate the acoustic properties of cardiovascular system and disorders of valves and thereby, in the future, could be used as a new diagnostic tool. Copyright © 2017. Published by Elsevier B.V.
Effects of de-oiled palm kernel cake based fertilizers on sole maize ...
A study was conducted to determine the effect of de-oiled palm kernel cake based fertilizer formulations on the yield of sole maize and cassava crops. Two de-oiled palm kernel cake based fertilizer formulations A and B were compounded from different proportions of de-oiled palm kernel cake, urea, muriate of potash and ...
Kernel based pattern analysis methods using eigen-decompositions for reading Icelandic sagas
Christiansen, Asger Nyman; Carstensen, Jens Michael
We want to test the applicability of kernel based eigen-decomposition methods, compared to the traditional eigen-decomposition methods. We have implemented and tested three kernel based methods methods, namely PCA, MAF and MNF, all using a Gaussian kernel. We tested the methods on a multispectral...... image of a page in the book 'hauksbok', which contains Icelandic sagas....
Geometric mean for subspace selection.
Tao, Dacheng; Li, Xuelong; Wu, Xindong; Maybank, Stephen J
2009-02-01
Subspace selection approaches are powerful tools in pattern classification and data visualization. One of the most important subspace approaches is the linear dimensionality reduction step in the Fisher's linear discriminant analysis (FLDA), which has been successfully employed in many fields such as biometrics, bioinformatics, and multimedia information management. However, the linear dimensionality reduction step in FLDA has a critical drawback: for a classification task with c classes, if the dimension of the projected subspace is strictly lower than c - 1, the projection to a subspace tends to merge those classes, which are close together in the original feature space. If separate classes are sampled from Gaussian distributions, all with identical covariance matrices, then the linear dimensionality reduction step in FLDA maximizes the mean value of the Kullback-Leibler (KL) divergences between different classes. Based on this viewpoint, the geometric mean for subspace selection is studied in this paper. Three criteria are analyzed: 1) maximization of the geometric mean of the KL divergences, 2) maximization of the geometric mean of the normalized KL divergences, and 3) the combination of 1 and 2. Preliminary experimental results based on synthetic data, UCI Machine Learning Repository, and handwriting digits show that the third criterion is a potential discriminative subspace selection method, which significantly reduces the class separation problem in comparing with the linear dimensionality reduction step in FLDA and its several representative extensions.
A multi-label learning based kernel automatic recommendation method for support vector machine.
Zhang, Xueying; Song, Qinbao
2015-01-01
Choosing an appropriate kernel is very important and critical when classifying a new problem with Support Vector Machine. So far, more attention has been paid on constructing new kernels and choosing suitable parameter values for a specific kernel function, but less on kernel selection. Furthermore, most of current kernel selection methods focus on seeking a best kernel with the highest classification accuracy via cross-validation, they are time consuming and ignore the differences among the number of support vectors and the CPU time of SVM with different kernels. Considering the tradeoff between classification success ratio and CPU time, there may be multiple kernel functions performing equally well on the same classification problem. Aiming to automatically select those appropriate kernel functions for a given data set, we propose a multi-label learning based kernel recommendation method built on the data characteristics. For each data set, the meta-knowledge data base is first created by extracting the feature vector of data characteristics and identifying the corresponding applicable kernel set. Then the kernel recommendation model is constructed on the generated meta-knowledge data base with the multi-label classification method. Finally, the appropriate kernel functions are recommended to a new data set by the recommendation model according to the characteristics of the new data set. Extensive experiments over 132 UCI benchmark data sets, with five different types of data set characteristics, eleven typical kernels (Linear, Polynomial, Radial Basis Function, Sigmoidal function, Laplace, Multiquadric, Rational Quadratic, Spherical, Spline, Wave and Circular), and five multi-label classification methods demonstrate that, compared with the existing kernel selection methods and the most widely used RBF kernel function, SVM with the kernel function recommended by our proposed method achieved the highest classification performance.
Müller, Emmanuel; Assent, Ira; Günnemann, Stephan
2009-01-01
Subspace clustering aims at detecting clusters in any subspace projection of a high dimensional space. As the number of possible subspace projections is exponential in the number of dimensions, the result is often tremendously large. Recent approaches fail to reduce results to relevant subspace...... clusters. Their results are typically highly redundant, i.e. many clusters are detected multiple times in several projections. In this work, we propose a novel model for relevant subspace clustering (RESCU). We present a global optimization which detects the most interesting non-redundant subspace clusters...... achieves top clustering quality while competing approaches show greatly varying performance....
Visual tracking based on the sparse representation of the PCA subspace
Chen, Dian-bing; Zhu, Ming; Wang, Hui-li
2017-09-01
We construct a collaborative model of the sparse representation and the subspace representation. First, we represent the tracking target in the principle component analysis (PCA) subspace, and then we employ an L 1 regularization to restrict the sparsity of the residual term, an L 2 regularization term to restrict the sparsity of the representation coefficients, and an L 2 norm to restrict the distance between the reconstruction and the target. Then we implement the algorithm in the particle filter framework. Furthermore, an iterative method is presented to get the global minimum of the residual and the coefficients. Finally, an alternative template update scheme is adopted to avoid the tracking drift which is caused by the inaccurate update. In the experiment, we test the algorithm on 9 sequences, and compare the results with 5 state-of-art methods. According to the results, we can conclude that our algorithm is more robust than the other methods.
Cumulant-Based Coherent Signal Subspace Method for Bearing and Range Estimation
Bourennane Salah
2007-01-01
Full Text Available A new method for simultaneous range and bearing estimation for buried objects in the presence of an unknown Gaussian noise is proposed. This method uses the MUSIC algorithm with noise subspace estimated by using the slice fourth-order cumulant matrix of the received data. The higher-order statistics aim at the removal of the additive unknown Gaussian noise. The bilinear focusing operator is used to decorrelate the received signals and to estimate the coherent signal subspace. A new source steering vector is proposed including the acoustic scattering model at each sensor. Range and bearing of the objects at each sensor are expressed as a function of those at the first sensor. This leads to the improvement of object localization anywhere, in the near-field or in the far-field zone of the sensor array. Finally, the performances of the proposed method are validated on data recorded during experiments in a water tank.
A Rank-Constrained Matrix Representation for Hypergraph-Based Subspace Clustering
Yubao Sun
2015-01-01
Full Text Available This paper presents a novel, rank-constrained matrix representation combined with hypergraph spectral analysis to enable the recovery of the original subspace structures of corrupted data. Real-world data are frequently corrupted with both sparse error and noise. Our matrix decomposition model separates the low-rank, sparse error, and noise components from the data in order to enhance robustness to the corruption. In order to obtain the desired rank representation of the data within a dictionary, our model directly utilizes rank constraints by restricting the upper bound of the rank range. An alternative projection algorithm is proposed to estimate the low-rank representation and separate the sparse error from the data matrix. To further capture the complex relationship between data distributed in multiple subspaces, we use hypergraph to represent the data by encapsulating multiple related samples into one hyperedge. The final clustering result is obtained by spectral decomposition of the hypergraph Laplacian matrix. Validation experiments on the Extended Yale Face Database B, AR, and Hopkins 155 datasets show that the proposed method is a promising tool for subspace clustering.
Chen, Tianshi; Andersen, Martin Skovgaard; Ljung, Lennart
2014-01-01
Model estimation and structure detection with short data records are two issues that receive increasing interests in System Identification. In this paper, a multiple kernel-based regularization method is proposed to handle those issues. Multiple kernels are conic combinations of fixed kernels...
Hailun Wang
2017-01-01
Full Text Available Support vector regression algorithm is widely used in fault diagnosis of rolling bearing. A new model parameter selection method for support vector regression based on adaptive fusion of the mixed kernel function is proposed in this paper. We choose the mixed kernel function as the kernel function of support vector regression. The mixed kernel function of the fusion coefficients, kernel function parameters, and regression parameters are combined together as the parameters of the state vector. Thus, the model selection problem is transformed into a nonlinear system state estimation problem. We use a 5th-degree cubature Kalman filter to estimate the parameters. In this way, we realize the adaptive selection of mixed kernel function weighted coefficients and the kernel parameters, the regression parameters. Compared with a single kernel function, unscented Kalman filter (UKF support vector regression algorithms, and genetic algorithms, the decision regression function obtained by the proposed method has better generalization ability and higher prediction accuracy.
Consistency Analysis of Nearest Subspace Classifier
Wang, Yi
2015-01-01
The Nearest subspace classifier (NSS) finds an estimation of the underlying subspace within each class and assigns data points to the class that corresponds to its nearest subspace. This paper mainly studies how well NSS can be generalized to new samples. It is proved that NSS is strongly consistent under certain assumptions. For completeness, NSS is evaluated through experiments on various simulated and real data sets, in comparison with some other linear model based classifiers. It is also ...
Nonlinear Knowledge in Kernel-Based Multiple Criteria Programming Classifier
Zhang, Dongling; Tian, Yingjie; Shi, Yong
Kernel-based Multiple Criteria Linear Programming (KMCLP) model is used as classification methods, which can learn from training examples. Whereas, in traditional machine learning area, data sets are classified only by prior knowledge. Some works combine the above two classification principle to overcome the defaults of each approach. In this paper, we propose a model to incorporate the nonlinear knowledge into KMCLP in order to solve the problem when input consists of not only training example, but also nonlinear prior knowledge. In dealing with real world case breast cancer diagnosis, the model shows its better performance than the model solely based on training data.
Random subspaces for encryption based on a private shared Cartesian frame
Bartlett, Stephen D.; Hayden, Patrick; Spekkens, Robert W.
2005-01-01
A private shared Cartesian frame is a novel form of private shared correlation that allows for both private classical and quantum communication. Cryptography using a private shared Cartesian frame has the remarkable property that asymptotically, if perfect privacy is demanded, the private classical capacity is three times the private quantum capacity. We demonstrate that if the requirement for perfect privacy is relaxed, then it is possible to use the properties of random subspaces to nearly triple the private quantum capacity, almost closing the gap between the private classical and quantum capacities
Towards automatic music transcription: note extraction based on independent subspace analysis
Wellhausen, Jens; Hoynck, Michael
2005-01-01
Due to the increasing amount of music available electronically the need of automatic search, retrieval and classification systems for music becomes more and more important. In this paper an algorithm for automatic transcription of polyphonic piano music into MIDI data is presented, which is a very interesting basis for database applications, music analysis and music classification. The first part of the algorithm performs a note accurate temporal audio segmentation. In the second part, the resulting segments are examined using Independent Subspace Analysis to extract sounding notes. Finally, the results are used to build a MIDI file as a new representation of the piece of music which is examined.
Speech Denoising in White Noise Based on Signal Subspace Low-rank Plus Sparse Decomposition
yuan Shuai
2017-01-01
Full Text Available In this paper, a new subspace speech enhancement method using low-rank and sparse decomposition is presented. In the proposed method, we firstly structure the corrupted data as a Toeplitz matrix and estimate its effective rank for the underlying human speech signal. Then the low-rank and sparse decomposition is performed with the guidance of speech rank value to remove the noise. Extensive experiments have been carried out in white Gaussian noise condition, and experimental results show the proposed method performs better than conventional speech enhancement methods, in terms of yielding less residual noise and lower speech distortion.
Müller, Emmanuel; Assent, Ira; Günnemann, Stephan
2009-01-01
Subspace clustering and projected clustering are recent research areas for clustering in high dimensional spaces. As the field is rather young, there is a lack of comparative studies on the advantages and disadvantages of the different algorithms. Part of the underlying problem is the lack...... of available open source implementations that could be used by researchers to understand, compare, and extend subspace and projected clustering algorithms. In this paper, we discuss the requirements for open source evaluation software. We propose OpenSubspace, an open source framework that meets...... these requirements. OpenSubspace integrates state-of-the-art performance measures and visualization techniques to foster research in subspace and projected clustering....
Abdel-Khalik, Hany S.; Zhang, Qiong
2014-01-01
The development of hybrid Monte-Carlo-Deterministic (MC-DT) approaches, taking place over the past few decades, have primarily focused on shielding and detection applications where the analysis requires a small number of responses, i.e. at the detector locations(s). This work further develops a recently introduced global variance reduction approach, denoted by the SUBSPACE approach is designed to allow the use of MC simulation, currently limited to benchmarking calculations, for routine engineering calculations. By way of demonstration, the SUBSPACE approach is applied to assembly level calculations used to generate the few-group homogenized cross-sections. These models are typically expensive and need to be executed in the order of 10 3 - 10 5 times to properly characterize the few-group cross-sections for downstream core-wide calculations. Applicability to k-eigenvalue core-wide models is also demonstrated in this work. Given the favorable results obtained in this work, we believe the applicability of the MC method for reactor analysis calculations could be realized in the near future.
Structural damage diagnosis based on on-line recursive stochastic subspace identification
Loh, Chin-Hsiung; Weng, Jian-Huang; Liu, Yi-Cheng; Lin, Pei-Yang; Huang, Shieh-Kung
2011-01-01
This paper presents a recursive stochastic subspace identification (RSSI) technique for on-line and almost real-time structural damage diagnosis using output-only measurements. Through RSSI the time-varying natural frequencies of a system can be identified. To reduce the computation time in conducting LQ decomposition in RSSI, the Givens rotation as well as the matrix operation appending a new data set are derived. The relationship between the size of the Hankel matrix and the data length in each shifting moving window is examined so as to extract the time-varying features of the system without loss of generality and to establish on-line and almost real-time system identification. The result from the RSSI technique can also be applied to structural damage diagnosis. Off-line data-driven stochastic subspace identification was used first to establish the system matrix from the measurements of an undamaged (reference) case. Then the RSSI technique incorporating a Kalman estimator is used to extract the dynamic characteristics of the system through continuous monitoring data. The predicted residual error is defined as a damage feature and through the outlier statistics provides an indicator of damage. Verification of the proposed identification algorithm by using the bridge scouring test data and white noise response data of a reinforced concrete frame structure is conducted
High resolution through-the-wall radar image based on beamspace eigenstructure subspace methods
Yoon, Yeo-Sun; Amin, Moeness G.
2008-04-01
Through-the-wall imaging (TWI) is a challenging problem, even if the wall parameters and characteristics are known to the system operator. Proper target classification and correct imaging interpretation require the application of high resolution techniques using limited array size. In inverse synthetic aperture radar (ISAR), signal subspace methods such as Multiple Signal Classification (MUSIC) are used to obtain high resolution imaging. In this paper, we adopt signal subspace methods and apply them to the 2-D spectrum obtained from the delay-andsum beamforming image. This is in contrast to ISAR, where raw data, in frequency and angle, is directly used to form the estimate of the covariance matrix and array response vector. Using beams rather than raw data has two main advantages, namely, it improves the signal-to-noise ratio (SNR) and can correctly image typical indoor extended targets, such as tables and cabinets, as well as point targets. The paper presents both simulated and experimental results using synthesized and real data. It compares the performance of beam-space MUSIC and Capon beamformer. The experimental data is collected at the test facility in the Radar Imaging Laboratory, Villanova University.
Efficient Kernel-Based Ensemble Gaussian Mixture Filtering
Liu, Bo
2015-11-11
We consider the Bayesian filtering problem for data assimilation following the kernel-based ensemble Gaussian-mixture filtering (EnGMF) approach introduced by Anderson and Anderson (1999). In this approach, the posterior distribution of the system state is propagated with the model using the ensemble Monte Carlo method, providing a forecast ensemble that is then used to construct a prior Gaussian-mixture (GM) based on the kernel density estimator. This results in two update steps: a Kalman filter (KF)-like update of the ensemble members and a particle filter (PF)-like update of the weights, followed by a resampling step to start a new forecast cycle. After formulating EnGMF for any observational operator, we analyze the influence of the bandwidth parameter of the kernel function on the covariance of the posterior distribution. We then focus on two aspects: i) the efficient implementation of EnGMF with (relatively) small ensembles, where we propose a new deterministic resampling strategy preserving the first two moments of the posterior GM to limit the sampling error; and ii) the analysis of the effect of the bandwidth parameter on contributions of KF and PF updates and on the weights variance. Numerical results using the Lorenz-96 model are presented to assess the behavior of EnGMF with deterministic resampling, study its sensitivity to different parameters and settings, and evaluate its performance against ensemble KFs. The proposed EnGMF approach with deterministic resampling suggests improved estimates in all tested scenarios, and is shown to require less localization and to be less sensitive to the choice of filtering parameters.
A class of kernel based real-time elastography algorithms.
Kibria, Md Golam; Hasan, Md Kamrul
2015-08-01
In this paper, a novel real-time kernel-based and gradient-based Phase Root Seeking (PRS) algorithm for ultrasound elastography is proposed. The signal-to-noise ratio of the strain image resulting from this method is improved by minimizing the cross-correlation discrepancy between the pre- and post-compression radio frequency signals with an adaptive temporal stretching method and employing built-in smoothing through an exponentially weighted neighborhood kernel in the displacement calculation. Unlike conventional PRS algorithms, displacement due to tissue compression is estimated from the root of the weighted average of the zero-lag cross-correlation phases of the pair of corresponding analytic pre- and post-compression windows in the neighborhood kernel. In addition to the proposed one, the other time- and frequency-domain elastography algorithms (Ara et al., 2013; Hussain et al., 2012; Hasan et al., 2012) proposed by our group are also implemented in real-time using Java where the computations are serially executed or parallely executed in multiple processors with efficient memory management. Simulation results using finite element modeling simulation phantom show that the proposed method significantly improves the strain image quality in terms of elastographic signal-to-noise ratio (SNRe), elastographic contrast-to-noise ratio (CNRe) and mean structural similarity (MSSIM) for strains as high as 4% as compared to other reported techniques in the literature. Strain images obtained for the experimental phantom as well as in vivo breast data of malignant or benign masses also show the efficacy of our proposed method over the other reported techniques in the literature. Copyright © 2015 Elsevier B.V. All rights reserved.
Kernel based orthogonalization for change detection in hyperspectral images
Nielsen, Allan Aasbjerg
function and all quantities needed in the analysis are expressed in terms of this kernel function. This means that we need not know the nonlinear mappings explicitly. Kernel PCA and MNF analyses handle nonlinearities by implicitly transforming data into high (even infinite) dimensional feature space via...... analysis all 126 spectral bands of the HyMap are included. Changes on the ground are most likely due to harvest having taken place between the two acquisitions and solar effects (both solar elevation and azimuth have changed). Both types of kernel analysis emphasize change and unlike kernel PCA, kernel MNF...
Kernel Based Nonlinear Dimensionality Reduction and Classification for Genomic Microarray
Lan Shu
2008-07-01
Full Text Available Genomic microarrays are powerful research tools in bioinformatics and modern medicinal research because they enable massively-parallel assays and simultaneous monitoring of thousands of gene expression of biological samples. However, a simple microarray experiment often leads to very high-dimensional data and a huge amount of information, the vast amount of data challenges researchers into extracting the important features and reducing the high dimensionality. In this paper, a nonlinear dimensionality reduction kernel method based locally linear embedding(LLE is proposed, and fuzzy K-nearest neighbors algorithm which denoises datasets will be introduced as a replacement to the classical LLEÃ¢Â€Â™s KNN algorithm. In addition, kernel method based support vector machine (SVM will be used to classify genomic microarray data sets in this paper. We demonstrate the application of the techniques to two published DNA microarray data sets. The experimental results confirm the superiority and high success rates of the presented method.
Osteoarthritis Severity Determination using Self Organizing Map Based Gabor Kernel
Anifah, L.; Purnomo, M. H.; Mengko, T. L. R.; Purnama, I. K. E.
2018-02-01
The number of osteoarthritis patients in Indonesia is enormous, so early action is needed in order for this disease to be handled. The aim of this paper to determine osteoarthritis severity based on x-ray image template based on gabor kernel. This research is divided into 3 stages, the first step is image processing that is using gabor kernel. The second stage is the learning stage, and the third stage is the testing phase. The image processing stage is by normalizing the image dimension to be template to 50 □ 200 image. Learning stage is done with parameters initial learning rate of 0.5 and the total number of iterations of 1000. The testing stage is performed using the weights generated at the learning stage. The testing phase has been done and the results were obtained. The result shows KL-Grade 0 has an accuracy of 36.21%, accuracy for KL-Grade 2 is 40,52%, while accuracy for KL-Grade 2 and KL-Grade 3 are 15,52%, and 25,86%. The implication of this research is expected that this research as decision support system for medical practitioners in determining KL-Grade on X-ray images of knee osteoarthritis.
Robust Pedestrian Classification Based on Hierarchical Kernel Sparse Representation
Rui Sun
2016-08-01
Full Text Available Vision-based pedestrian detection has become an active topic in computer vision and autonomous vehicles. It aims at detecting pedestrians appearing ahead of the vehicle using a camera so that autonomous vehicles can assess the danger and take action. Due to varied illumination and appearance, complex background and occlusion pedestrian detection in outdoor environments is a difficult problem. In this paper, we propose a novel hierarchical feature extraction and weighted kernel sparse representation model for pedestrian classification. Initially, hierarchical feature extraction based on a CENTRIST descriptor is used to capture discriminative structures. A max pooling operation is used to enhance the invariance of varying appearance. Then, a kernel sparse representation model is proposed to fully exploit the discrimination information embedded in the hierarchical local features, and a Gaussian weight function as the measure to effectively handle the occlusion in pedestrian images. Extensive experiments are conducted on benchmark databases, including INRIA, Daimler, an artificially generated dataset and a real occluded dataset, demonstrating the more robust performance of the proposed method compared to state-of-the-art pedestrian classification methods.
External Evaluation Measures for Subspace Clustering
Günnemann, Stephan; Färber, Ines; Müller, Emmanuel
2011-01-01
research area of subspace clustering. We formalize general quality criteria for subspace clustering measures not yet addressed in the literature. We compare the existing external evaluation methods based on these criteria and pinpoint limitations. We propose a novel external evaluation measure which meets...
Subspace learning from image gradient orientations
Tzimiropoulos, Georgios; Zafeiriou, Stefanos; Pantic, Maja
2012-01-01
We introduce the notion of subspace learning from image gradient orientations for appearance-based object recognition. As image data is typically noisy and noise is substantially different from Gaussian, traditional subspace learning from pixel intensities fails very often to estimate reliably the
Siddiqui, Bilal A.; El-Ferik, Sami; Abdelkader, Mohamed
2016-01-01
In this work, a cascade structure of a time-scale separated integral sliding mode and model predictive control is proposed as a viable alternative for fault-tolerant control. A multi-variable sliding mode control law is designed as the inner loop of the flight control system. Subspace identification is carried out on the aircraft in closed loop. The identified plant is then used for model predictive controllers in the outer loop. The overall control law demonstrates improved robustness to measurement noise, modeling uncertainties, multiple faults and severe wind turbulence and gusts. In addition, the flight control system employs filters and dead-zone nonlinear elements to reduce chattering and improve handling quality. Simulation results demonstrate the efficiency of the proposed controller using conventional fighter aircraft without control redundancy.
Siddiqui, Bilal A.
2016-07-26
In this work, a cascade structure of a time-scale separated integral sliding mode and model predictive control is proposed as a viable alternative for fault-tolerant control. A multi-variable sliding mode control law is designed as the inner loop of the flight control system. Subspace identification is carried out on the aircraft in closed loop. The identified plant is then used for model predictive controllers in the outer loop. The overall control law demonstrates improved robustness to measurement noise, modeling uncertainties, multiple faults and severe wind turbulence and gusts. In addition, the flight control system employs filters and dead-zone nonlinear elements to reduce chattering and improve handling quality. Simulation results demonstrate the efficiency of the proposed controller using conventional fighter aircraft without control redundancy.
Shape analysis with subspace symmetries
Berner, Alexander
2011-04-01
We address the problem of partial symmetry detection, i.e., the identification of building blocks a complex shape is composed of. Previous techniques identify parts that relate to each other by simple rigid mappings, similarity transforms, or, more recently, intrinsic isometries. Our approach generalizes the notion of partial symmetries to more general deformations. We introduce subspace symmetries whereby we characterize similarity by requiring the set of symmetric parts to form a low dimensional shape space. We present an algorithm to discover subspace symmetries based on detecting linearly correlated correspondences among graphs of invariant features. We evaluate our technique on various data sets. We show that for models with pronounced surface features, subspace symmetries can be found fully automatically. For complicated cases, a small amount of user input is used to resolve ambiguities. Our technique computes dense correspondences that can subsequently be used in various applications, such as model repair and denoising. © 2010 The Author(s).
Alumina Concentration Detection Based on the Kernel Extreme Learning Machine.
Zhang, Sen; Zhang, Tao; Yin, Yixin; Xiao, Wendong
2017-09-01
The concentration of alumina in the electrolyte is of great significance during the production of aluminum. The amount of the alumina concentration may lead to unbalanced material distribution and low production efficiency and affect the stability of the aluminum reduction cell and current efficiency. The existing methods cannot meet the needs for online measurement because industrial aluminum electrolysis has the characteristics of high temperature, strong magnetic field, coupled parameters, and high nonlinearity. Currently, there are no sensors or equipment that can detect the alumina concentration on line. Most companies acquire the alumina concentration from the electrolyte samples which are analyzed through an X-ray fluorescence spectrometer. To solve the problem, the paper proposes a soft sensing model based on a kernel extreme learning machine algorithm that takes the kernel function into the extreme learning machine. K-fold cross validation is used to estimate the generalization error. The proposed soft sensing algorithm can detect alumina concentration by the electrical signals such as voltages and currents of the anode rods. The predicted results show that the proposed approach can give more accurate estimations of alumina concentration with faster learning speed compared with the other methods such as the basic ELM, BP, and SVM.
A trace ratio maximization approach to multiple kernel-based dimensionality reduction.
Jiang, Wenhao; Chung, Fu-lai
2014-01-01
Most dimensionality reduction techniques are based on one metric or one kernel, hence it is necessary to select an appropriate kernel for kernel-based dimensionality reduction. Multiple kernel learning for dimensionality reduction (MKL-DR) has been recently proposed to learn a kernel from a set of base kernels which are seen as different descriptions of data. As MKL-DR does not involve regularization, it might be ill-posed under some conditions and consequently its applications are hindered. This paper proposes a multiple kernel learning framework for dimensionality reduction based on regularized trace ratio, termed as MKL-TR. Our method aims at learning a transformation into a space of lower dimension and a corresponding kernel from the given base kernels among which some may not be suitable for the given data. The solutions for the proposed framework can be found based on trace ratio maximization. The experimental results demonstrate its effectiveness in benchmark datasets, which include text, image and sound datasets, for supervised, unsupervised as well as semi-supervised settings. Copyright © 2013 Elsevier Ltd. All rights reserved.
Sensitivity kernels for viscoelastic loading based on adjoint methods
Al-Attar, David; Tromp, Jeroen
2014-01-01
Observations of glacial isostatic adjustment (GIA) allow for inferences to be made about mantle viscosity, ice sheet history and other related parameters. Typically, this inverse problem can be formulated as minimizing the misfit between the given observations and a corresponding set of synthetic data. When the number of parameters is large, solution of such optimization problems can be computationally challenging. A practical, albeit non-ideal, solution is to use gradient-based optimization. Although the gradient of the misfit required in such methods could be calculated approximately using finite differences, the necessary computation time grows linearly with the number of model parameters, and so this is often infeasible. A far better approach is to apply the `adjoint method', which allows the exact gradient to be calculated from a single solution of the forward problem, along with one solution of the associated adjoint problem. As a first step towards applying the adjoint method to the GIA inverse problem, we consider its application to a simpler viscoelastic loading problem in which gravitationally self-consistent ocean loading is neglected. The earth model considered is non-rotating, self-gravitating, compressible, hydrostatically pre-stressed, laterally heterogeneous and possesses a Maxwell solid rheology. We determine adjoint equations and Fréchet kernels for this problem based on a Lagrange multiplier method. Given an objective functional J defined in terms of the surface deformation fields, we show that its first-order perturbation can be written δ J = int _{MS}K_{η }δ ln η dV +int _{t0}^{t1}int _{partial M}K_{dot{σ }} δ dot{σ } dS dt, where δ ln η = δη/η denotes relative viscosity variations in solid regions MS, dV is the volume element, δ dot{σ } is the perturbation to the time derivative of the surface load which is defined on the earth model's surface ∂M and for times [t0, t1] and dS is the surface element on ∂M. The `viscosity
Zarifi, Keyvan; Gershman, Alex B.
2006-12-01
We analyze the performance of two popular blind subspace-based signature waveform estimation techniques proposed by Wang and Poor and Buzzi and Poor for direct-sequence code division multiple-access (DS-CDMA) systems with unknown correlated noise. Using the first-order perturbation theory, analytical expressions for the mean-square error (MSE) of these algorithms are derived. We also obtain simple high SNR approximations of the MSE expressions which explicitly clarify how the performance of these techniques depends on the environmental parameters and how it is related to that of the conventional techniques that are based on the standard white noise assumption. Numerical examples further verify the consistency of the obtained analytical results with simulation results.
Xianglin ZHU
2014-06-01
Full Text Available To resolve the online detection difficulty of some important state variables in fermentation process with traditional instruments, a soft sensing modeling method based on relevance vector machine (RVM with a hybrid kernel function is presented. Based on the characteristic analysis of two commonly-used kernel functions, that is, local Gaussian kernel function and global polynomial kernel function, a hybrid kernel function combing merits of Gaussian kernel function and polynomial kernel function is constructed. To design optimal parameters of this kernel function, the particle swarm optimization (PSO algorithm is applied. The proposed modeling method is used to predict the value of cell concentration in the Lysine fermentation process. Simulation results show that the presented hybrid-kernel RVM model has a better accuracy and performance than the single kernel RVM model.
A framework for optimal kernel-based manifold embedding of medical image data.
Zimmer, Veronika A; Lekadir, Karim; Hoogendoorn, Corné; Frangi, Alejandro F; Piella, Gemma
2015-04-01
Kernel-based dimensionality reduction is a widely used technique in medical image analysis. To fully unravel the underlying nonlinear manifold the selection of an adequate kernel function and of its free parameters is critical. In practice, however, the kernel function is generally chosen as Gaussian or polynomial and such standard kernels might not always be optimal for a given image dataset or application. In this paper, we present a study on the effect of the kernel functions in nonlinear manifold embedding of medical image data. To this end, we first carry out a literature review on existing advanced kernels developed in the statistics, machine learning, and signal processing communities. In addition, we implement kernel-based formulations of well-known nonlinear dimensional reduction techniques such as Isomap and Locally Linear Embedding, thus obtaining a unified framework for manifold embedding using kernels. Subsequently, we present a method to automatically choose a kernel function and its associated parameters from a pool of kernel candidates, with the aim to generate the most optimal manifold embeddings. Furthermore, we show how the calculated selection measures can be extended to take into account the spatial relationships in images, or used to combine several kernels to further improve the embedding results. Experiments are then carried out on various synthetic and phantom datasets for numerical assessment of the methods. Furthermore, the workflow is applied to real data that include brain manifolds and multispectral images to demonstrate the importance of the kernel selection in the analysis of high-dimensional medical images. Copyright © 2014 Elsevier Ltd. All rights reserved.
An acceleration technique for 2D MOC based on Krylov subspace and domain decomposition methods
Zhang Hongbo; Wu Hongchun; Cao Liangzhi
2011-01-01
Highlights: → We convert MOC into linear system solved by GMRES as an acceleration method. → We use domain decomposition method to overcome the inefficiency on large matrices. → Parallel technology is applied and a matched ray tracing system is developed. → Results show good efficiency even in large-scale and strong scattering problems. → The emphasis is that the technique is geometry-flexible. - Abstract: The method of characteristics (MOC) has great geometrical flexibility but poor computational efficiency in neutron transport calculations. The generalized minimal residual (GMRES) method, a type of Krylov subspace method, is utilized to accelerate a 2D generalized geometry characteristics solver AutoMOC. In this technique, a form of linear algebraic equation system for angular flux moments and boundary fluxes is derived to replace the conventional characteristics sweep (i.e. inner iteration) scheme, and then the GMRES method is implemented as an efficient linear system solver. This acceleration method is proved to be reliable in theory and simple for implementation. Furthermore, as introducing no restriction in geometry treatment, it is suitable for acceleration of an arbitrary geometry MOC solver. However, it is observed that the speedup decreases when the matrix becomes larger. The spatial domain decomposition method and multiprocessing parallel technology are then employed to overcome the problem. The calculation domain is partitioned into several sub-domains. For each of them, a smaller matrix is established and solved by GMRES; and the adjacent sub-domains are coupled by 'inner-edges', where the trajectory mismatches are considered adequately. Moreover, a matched ray tracing system is developed on the basis of AutoCAD, which allows a user to define the sub-domains on demand conveniently. Numerical results demonstrate that the acceleration techniques are efficient without loss of accuracy, even in the case of large-scale and strong scattering
Jian, Yulin; Huang, Daoyu; Yan, Jia; Lu, Kun; Huang, Ying; Wen, Tailai; Zeng, Tanyue; Zhong, Shijie; Xie, Qilong
2017-06-19
A novel classification model, named the quantum-behaved particle swarm optimization (QPSO)-based weighted multiple kernel extreme learning machine (QWMK-ELM), is proposed in this paper. Experimental validation is carried out with two different electronic nose (e-nose) datasets. Being different from the existing multiple kernel extreme learning machine (MK-ELM) algorithms, the combination coefficients of base kernels are regarded as external parameters of single-hidden layer feedforward neural networks (SLFNs). The combination coefficients of base kernels, the model parameters of each base kernel, and the regularization parameter are optimized by QPSO simultaneously before implementing the kernel extreme learning machine (KELM) with the composite kernel function. Four types of common single kernel functions (Gaussian kernel, polynomial kernel, sigmoid kernel, and wavelet kernel) are utilized to constitute different composite kernel functions. Moreover, the method is also compared with other existing classification methods: extreme learning machine (ELM), kernel extreme learning machine (KELM), k-nearest neighbors (KNN), support vector machine (SVM), multi-layer perceptron (MLP), radical basis function neural network (RBFNN), and probabilistic neural network (PNN). The results have demonstrated that the proposed QWMK-ELM outperforms the aforementioned methods, not only in precision, but also in efficiency for gas classification.
Fuzzy-based multi-kernel spherical support vector machine for ...
In the proposed classifier, we design a new multi-kernel function based on the fuzzy triangular membership function. Finally, a newly developed multi-kernel function is incorporated into the spherical support vector machine to enhance the performance significantly. The experimental results are evaluated and performance is ...
Pathway-Based Kernel Boosting for the Analysis of Genome-Wide Association Studies
Manitz, Juliane; Burger, Patricia; Amos, Christopher I.; Chang-Claude, Jenny; Wichmann, Heinz-Erich; Kneib, Thomas; Bickeböller, Heike
2017-01-01
The analysis of genome-wide association studies (GWAS) benefits from the investigation of biologically meaningful gene sets, such as gene-interaction networks (pathways). We propose an extension to a successful kernel-based pathway analysis approach by integrating kernel functions into a powerful algorithmic framework for variable selection, to enable investigation of multiple pathways simultaneously. We employ genetic similarity kernels from the logistic kernel machine test (LKMT) as base-learners in a boosting algorithm. A model to explain case-control status is created iteratively by selecting pathways that improve its prediction ability. We evaluated our method in simulation studies adopting 50 pathways for different sample sizes and genetic effect strengths. Additionally, we included an exemplary application of kernel boosting to a rheumatoid arthritis and a lung cancer dataset. Simulations indicate that kernel boosting outperforms the LKMT in certain genetic scenarios. Applications to GWAS data on rheumatoid arthritis and lung cancer resulted in sparse models which were based on pathways interpretable in a clinical sense. Kernel boosting is highly flexible in terms of considered variables and overcomes the problem of multiple testing. Additionally, it enables the prediction of clinical outcomes. Thus, kernel boosting constitutes a new, powerful tool in the analysis of GWAS data and towards the understanding of biological processes involved in disease susceptibility. PMID:28785300
Pathway-Based Kernel Boosting for the Analysis of Genome-Wide Association Studies.
Friedrichs, Stefanie; Manitz, Juliane; Burger, Patricia; Amos, Christopher I; Risch, Angela; Chang-Claude, Jenny; Wichmann, Heinz-Erich; Kneib, Thomas; Bickeböller, Heike; Hofner, Benjamin
2017-01-01
The analysis of genome-wide association studies (GWAS) benefits from the investigation of biologically meaningful gene sets, such as gene-interaction networks (pathways). We propose an extension to a successful kernel-based pathway analysis approach by integrating kernel functions into a powerful algorithmic framework for variable selection, to enable investigation of multiple pathways simultaneously. We employ genetic similarity kernels from the logistic kernel machine test (LKMT) as base-learners in a boosting algorithm. A model to explain case-control status is created iteratively by selecting pathways that improve its prediction ability. We evaluated our method in simulation studies adopting 50 pathways for different sample sizes and genetic effect strengths. Additionally, we included an exemplary application of kernel boosting to a rheumatoid arthritis and a lung cancer dataset. Simulations indicate that kernel boosting outperforms the LKMT in certain genetic scenarios. Applications to GWAS data on rheumatoid arthritis and lung cancer resulted in sparse models which were based on pathways interpretable in a clinical sense. Kernel boosting is highly flexible in terms of considered variables and overcomes the problem of multiple testing. Additionally, it enables the prediction of clinical outcomes. Thus, kernel boosting constitutes a new, powerful tool in the analysis of GWAS data and towards the understanding of biological processes involved in disease susceptibility.
Characterisation and final disposal behaviour of theoria-based fuel kernels in aqueous phases
Titov, M.
2005-08-01
Two high-temperature reactors (AVR and THTR) operated in Germany have produced about 1 million spent fuel elements. The nuclear fuel in these reactors consists mainly of thorium-uranium mixed oxides, but also pure uranium dioxide and carbide fuels were tested. One of the possible solutions of utilising spent HTR fuel is the direct disposal in deep geological formations. Under such circumstances, the properties of fuel kernels, and especially their leaching behaviour in aqueous phases, have to be investigated for safety assessments of the final repository. In the present work, unirradiated ThO 2 , (Th 0.906 ,U 0.094 )O 2 , (Th 0.834 ,U 0.166 )O 2 and UO 2 fuel kernels were investigated. The composition, crystal structure and surface of the kernels were investigated by traditional methods. Furthermore, a new method was developed for testing the mechanical properties of ceramic kernels. The method was successfully used for the examination of mechanical properties of oxide kernels and for monitoring their evolution during contact with aqueous phases. The leaching behaviour of thoria-based oxide kernels and powders was investigated in repository-relevant salt solutions, as well as in artificial leachates. The influence of different experimental parameters on the kernel leaching stability was investigated. It was shown that thoria-based fuel kernels possess high chemical stability and are indifferent to presence of oxidative and radiolytic species in solution. The dissolution rate of thoria-based materials is typically several orders of magnitude lower than of conventional UO 2 fuel kernels. The life time of a single intact (Th,U)O 2 kernel under aggressive conditions of salt repository was estimated as about hundred thousand years. The importance of grain boundary quality on the leaching stability was demonstrated. Numerical Monte Carlo simulations were performed in order to explain the results of leaching experiments. (orig.)
Fuzzy-based multi-kernel spherical support vector machine for ...
A K Sampath
2017-08-08
Aug 8, 2017 ... design a new multi-kernel function based on the fuzzy triangular membership function. Finally .... This paper is structured as follows. Section 2 ..... analysis is compared with some existing systems based on the number of ...
Kernel-based discriminant feature extraction using a representative dataset
Li, Honglin; Sancho Gomez, Jose-Luis; Ahalt, Stanley C.
2002-07-01
Discriminant Feature Extraction (DFE) is widely recognized as an important pre-processing step in classification applications. Most DFE algorithms are linear and thus can only explore the linear discriminant information among the different classes. Recently, there has been several promising attempts to develop nonlinear DFE algorithms, among which is Kernel-based Feature Extraction (KFE). The efficacy of KFE has been experimentally verified by both synthetic data and real problems. However, KFE has some known limitations. First, KFE does not work well for strongly overlapped data. Second, KFE employs all of the training set samples during the feature extraction phase, which can result in significant computation when applied to very large datasets. Finally, KFE can result in overfitting. In this paper, we propose a substantial improvement to KFE that overcomes the above limitations by using a representative dataset, which consists of critical points that are generated from data-editing techniques and centroid points that are determined by using the Frequency Sensitive Competitive Learning (FSCL) algorithm. Experiments show that this new KFE algorithm performs well on significantly overlapped datasets, and it also reduces computational complexity. Further, by controlling the number of centroids, the overfitting problem can be effectively alleviated.
LAI inversion algorithm based on directional reflectance kernels.
Tang, S; Chen, J M; Zhu, Q; Li, X; Chen, M; Sun, R; Zhou, Y; Deng, F; Xie, D
2007-11-01
Leaf area index (LAI) is an important ecological and environmental parameter. A new LAI algorithm is developed using the principles of ground LAI measurements based on canopy gap fraction. First, the relationship between LAI and gap fraction at various zenith angles is derived from the definition of LAI. Then, the directional gap fraction is acquired from a remote sensing bidirectional reflectance distribution function (BRDF) product. This acquisition is obtained by using a kernel driven model and a large-scale directional gap fraction algorithm. The algorithm has been applied to estimate a LAI distribution in China in mid-July 2002. The ground data acquired from two field experiments in Changbai Mountain and Qilian Mountain were used to validate the algorithm. To resolve the scale discrepancy between high resolution ground observations and low resolution remote sensing data, two TM images with a resolution approaching the size of ground plots were used to relate the coarse resolution LAI map to ground measurements. First, an empirical relationship between the measured LAI and a vegetation index was established. Next, a high resolution LAI map was generated using the relationship. The LAI value of a low resolution pixel was calculated from the area-weighted sum of high resolution LAIs composing the low resolution pixel. The results of this comparison showed that the inversion algorithm has an accuracy of 82%. Factors that may influence the accuracy are also discussed in this paper.
Gradient descent for robust kernel-based regression
Guo, Zheng-Chu; Hu, Ting; Shi, Lei
2018-06-01
In this paper, we study the gradient descent algorithm generated by a robust loss function over a reproducing kernel Hilbert space (RKHS). The loss function is defined by a windowing function G and a scale parameter σ, which can include a wide range of commonly used robust losses for regression. There is still a gap between theoretical analysis and optimization process of empirical risk minimization based on loss: the estimator needs to be global optimal in the theoretical analysis while the optimization method can not ensure the global optimality of its solutions. In this paper, we aim to fill this gap by developing a novel theoretical analysis on the performance of estimators generated by the gradient descent algorithm. We demonstrate that with an appropriately chosen scale parameter σ, the gradient update with early stopping rules can approximate the regression function. Our elegant error analysis can lead to convergence in the standard L 2 norm and the strong RKHS norm, both of which are optimal in the mini-max sense. We show that the scale parameter σ plays an important role in providing robustness as well as fast convergence. The numerical experiments implemented on synthetic examples and real data set also support our theoretical results.
Yi-Hung Liu
2014-01-01
Full Text Available In this paper, we propose a robust tactile sensing image recognition scheme for automatic robotic assembly. First, an image reprocessing procedure is designed to enhance the contrast of the tactile image. In the second layer, geometric features and Fourier descriptors are extracted from the image. Then, kernel principal component analysis (kernel PCA is applied to transform the features into ones with better discriminating ability, which is the kernel PCA-based feature fusion. The transformed features are fed into the third layer for classification. In this paper, we design a classifier by combining the multiple kernel learning (MKL algorithm and support vector machine (SVM. We also design and implement a tactile sensing array consisting of 10-by-10 sensing elements. Experimental results, carried out on real tactile images acquired by the designed tactile sensing array, show that the kernel PCA-based feature fusion can significantly improve the discriminating performance of the geometric features and Fourier descriptors. Also, the designed MKL-SVM outperforms the regular SVM in terms of recognition accuracy. The proposed recognition scheme is able to achieve a high recognition rate of over 85% for the classification of 12 commonly used metal parts in industrial applications.
Zhang, Liangwei; Lin, Jing; Karim, Ramin
2015-01-01
The accuracy of traditional anomaly detection techniques implemented on full-dimensional spaces degrades significantly as dimensionality increases, thereby hampering many real-world applications. This work proposes an approach to selecting meaningful feature subspace and conducting anomaly detection in the corresponding subspace projection. The aim is to maintain the detection accuracy in high-dimensional circumstances. The suggested approach assesses the angle between all pairs of two lines for one specific anomaly candidate: the first line is connected by the relevant data point and the center of its adjacent points; the other line is one of the axis-parallel lines. Those dimensions which have a relatively small angle with the first line are then chosen to constitute the axis-parallel subspace for the candidate. Next, a normalized Mahalanobis distance is introduced to measure the local outlier-ness of an object in the subspace projection. To comprehensively compare the proposed algorithm with several existing anomaly detection techniques, we constructed artificial datasets with various high-dimensional settings and found the algorithm displayed superior accuracy. A further experiment on an industrial dataset demonstrated the applicability of the proposed algorithm in fault detection tasks and highlighted another of its merits, namely, to provide preliminary interpretation of abnormality through feature ordering in relevant subspaces. - Highlights: • An anomaly detection approach for high-dimensional reliability data is proposed. • The approach selects relevant subspaces by assessing vectorial angles. • The novel ABSAD approach displays superior accuracy over other alternatives. • Numerical illustration approves its efficacy in fault detection applications
An OSKit-Based Implementation of Least Privilege Separation Kernel Memory Partitioning
Carter, Donald W
2007-01-01
.... This work is to build a working prototype of selected TCX kernel functionality. The prototype is constructed and based on OSKit, and restricts information flow between memory partitions and resource accesses...
Music recommendation according to human motion based on kernel CCA-based relationship
Ohkushi, Hiroyuki; Ogawa, Takahiro; Haseyama, Miki
2011-12-01
In this article, a method for recommendation of music pieces according to human motions based on their kernel canonical correlation analysis (CCA)-based relationship is proposed. In order to perform the recommendation between different types of multimedia data, i.e., recommendation of music pieces from human motions, the proposed method tries to estimate their relationship. Specifically, the correlation based on kernel CCA is calculated as the relationship in our method. Since human motions and music pieces have various time lengths, it is necessary to calculate the correlation between time series having different lengths. Therefore, new kernel functions for human motions and music pieces, which can provide similarities between data that have different time lengths, are introduced into the calculation of the kernel CCA-based correlation. This approach effectively provides a solution to the conventional problem of not being able to calculate the correlation from multimedia data that have various time lengths. Therefore, the proposed method can perform accurate recommendation of best matched music pieces according to a target human motion from the obtained correlation. Experimental results are shown to verify the performance of the proposed method.
Sinusoidal Order Estimation Using Angles between Subspaces
Søren Holdt Jensen
2009-01-01
Full Text Available We consider the problem of determining the order of a parametric model from a noisy signal based on the geometry of the space. More specifically, we do this using the nontrivial angles between the candidate signal subspace model and the noise subspace. The proposed principle is closely related to the subspace orthogonality property known from the MUSIC algorithm, and we study its properties and compare it to other related measures. For the problem of estimating the number of complex sinusoids in white noise, a computationally efficient implementation exists, and this problem is therefore considered in detail. In computer simulations, we compare the proposed method to various well-known methods for order estimation. These show that the proposed method outperforms the other previously published subspace methods and that it is more robust to the noise being colored than the previously published methods.
A Walk-based Semantically Enriched Tree Kernel Over Distributed Word Representations
Srivastava, Shashank; Hovy, Dirk
2013-01-01
We propose a walk-based graph kernel that generalizes the notion of tree-kernels to continuous spaces. Our proposed approach subsumes a general framework for word-similarity, and in particular, provides a flexible way to incorporate distributed representations. Using vector representations......, such an approach captures both distributional semantic similarities among words as well as the structural relations between them (encoded as the structure of the parse tree). We show an efficient formulation to compute this kernel using simple matrix multiplication operations. We present our results on three...
Adaptive Shape Kernel-Based Mean Shift Tracker in Robot Vision System
Chunmei Liu
2016-01-01
Full Text Available This paper proposes an adaptive shape kernel-based mean shift tracker using a single static camera for the robot vision system. The question that we address in this paper is how to construct such a kernel shape that is adaptive to the object shape. We perform nonlinear manifold learning technique to obtain the low-dimensional shape space which is trained by training data with the same view as the tracking video. The proposed kernel searches the shape in the low-dimensional shape space obtained by nonlinear manifold learning technique and constructs the adaptive kernel shape in the high-dimensional shape space. It can improve mean shift tracker performance to track object position and object contour and avoid the background clutter. In the experimental part, we take the walking human as example to validate that our method is accurate and robust to track human position and describe human contour.
Adaptive Shape Kernel-Based Mean Shift Tracker in Robot Vision System
2016-01-01
This paper proposes an adaptive shape kernel-based mean shift tracker using a single static camera for the robot vision system. The question that we address in this paper is how to construct such a kernel shape that is adaptive to the object shape. We perform nonlinear manifold learning technique to obtain the low-dimensional shape space which is trained by training data with the same view as the tracking video. The proposed kernel searches the shape in the low-dimensional shape space obtained by nonlinear manifold learning technique and constructs the adaptive kernel shape in the high-dimensional shape space. It can improve mean shift tracker performance to track object position and object contour and avoid the background clutter. In the experimental part, we take the walking human as example to validate that our method is accurate and robust to track human position and describe human contour. PMID:27379165
On Covering Approximation Subspaces
Xun Ge
2009-06-01
Full Text Available Let (U';C' be a subspace of a covering approximation space (U;C and X⊂U'. In this paper, we show that and B'(X⊂B(X∩U'. Also, iff (U;C has Property Multiplication. Furthermore, some connections between outer (resp. inner definable subsets in (U;C and outer (resp. inner definable subsets in (U';C' are established. These results answer a question on covering approximation subspace posed by J. Li, and are helpful to obtain further applications of Pawlak rough set theory in pattern recognition and artificial intelligence.
Carvajal, Gonzalo; Figueroa, Miguel
2014-07-01
Typical image recognition systems operate in two stages: feature extraction to reduce the dimensionality of the input space, and classification based on the extracted features. Analog Very Large Scale Integration (VLSI) is an attractive technology to achieve compact and low-power implementations of these computationally intensive tasks for portable embedded devices. However, device mismatch limits the resolution of the circuits fabricated with this technology. Traditional layout techniques to reduce the mismatch aim to increase the resolution at the transistor level, without considering the intended application. Relating mismatch parameters to specific effects in the application level would allow designers to apply focalized mismatch compensation techniques according to predefined performance/cost tradeoffs. This paper models, analyzes, and evaluates the effects of mismatched analog arithmetic in both feature extraction and classification circuits. For the feature extraction, we propose analog adaptive linear combiners with on-chip learning for both Least Mean Square (LMS) and Generalized Hebbian Algorithm (GHA). Using mathematical abstractions of analog circuits, we identify mismatch parameters that are naturally compensated during the learning process, and propose cost-effective guidelines to reduce the effect of the rest. For the classification, we derive analog models for the circuits necessary to implement Nearest Neighbor (NN) approach and Radial Basis Function (RBF) networks, and use them to emulate analog classifiers with standard databases of face and hand-writing digits. Formal analysis and experiments show how we can exploit adaptive structures and properties of the input space to compensate the effects of device mismatch at the application level, thus reducing the design overhead of traditional layout techniques. Results are also directly extensible to multiple application domains using linear subspace methods. Copyright © 2014 Elsevier Ltd. All rights
de la Torre Vega, E. [Instituto de Investigaciones Electricas, Cuernavaca (Mexico); Cesar Suarez Arriaga, M. [Universidad Michoacana SNH, Michoacan (Mexico)
1995-03-01
In geothermal simulation processes, MULKOM uses Integrated Finite Differences to solve the corresponding partial differential equations. This method requires to resolve efficiently big linear dispersed systems of non-symmetrical nature on each temporal iteration. The order of the system is usually greater than one thousand its solution could represent around 80% of CPU total calculation time. If the elapsed time solving this class of linear systems is reduced, the duration of numerical simulation decreases notably. When the matrix is big (N{ge}500) and with holes, it is inefficient to handle all the system`s elements, because it is perfectly figured out by its elements distinct of zero, quantity greatly minor than N{sup 2}. In this area, iteration methods introduce advantages with respect to gaussian elimination methods, because these last replenish matrices not having any special distribution of their non-zero elements and because they do not make use of the available solution estimations. The iterating methods of the Conjugated Gradient family, based on the subspaces of Krylov, possess the advantage of improving the convergence speed by means of preconditioning techniques. The creation of DIOMRES(k,m) method guarantees the continuous descent of the residual norm, without incurring in division by zero. This technique converges at most in N iterations if the system`s matrix is symmetrical, it does not employ too much memory to converge and updates immediately the approximation by using incomplete orthogonalization and adequate restarting. A preconditioned version of DIOMRES was applied to problems related to unsymmetrical systems with 1000 unknowns and less than five terms per equation. We found that this technique could reduce notably the time needful to find the solution without requiring memory increment. The coupling of this method to geothermal versions of MULKOM is in process.
Cai, Jia; Tang, Yi
2018-02-01
Canonical correlation analysis (CCA) is a powerful statistical tool for detecting the linear relationship between two sets of multivariate variables. Kernel generalization of it, namely, kernel CCA is proposed to describe nonlinear relationship between two variables. Although kernel CCA can achieve dimensionality reduction results for high-dimensional data feature selection problem, it also yields the so called over-fitting phenomenon. In this paper, we consider a new kernel CCA algorithm via randomized Kaczmarz method. The main contributions of the paper are: (1) A new kernel CCA algorithm is developed, (2) theoretical convergence of the proposed algorithm is addressed by means of scaled condition number, (3) a lower bound which addresses the minimum number of iterations is presented. We test on both synthetic dataset and several real-world datasets in cross-language document retrieval and content-based image retrieval to demonstrate the effectiveness of the proposed algorithm. Numerical results imply the performance and efficiency of the new algorithm, which is competitive with several state-of-the-art kernel CCA methods. Copyright © 2017 Elsevier Ltd. All rights reserved.
A Comparative Study of Pairwise Learning Methods Based on Kernel Ridge Regression.
Stock, Michiel; Pahikkala, Tapio; Airola, Antti; De Baets, Bernard; Waegeman, Willem
2018-06-12
Many machine learning problems can be formulated as predicting labels for a pair of objects. Problems of that kind are often referred to as pairwise learning, dyadic prediction, or network inference problems. During the past decade, kernel methods have played a dominant role in pairwise learning. They still obtain a state-of-the-art predictive performance, but a theoretical analysis of their behavior has been underexplored in the machine learning literature. In this work we review and unify kernel-based algorithms that are commonly used in different pairwise learning settings, ranging from matrix filtering to zero-shot learning. To this end, we focus on closed-form efficient instantiations of Kronecker kernel ridge regression. We show that independent task kernel ridge regression, two-step kernel ridge regression, and a linear matrix filter arise naturally as a special case of Kronecker kernel ridge regression, implying that all these methods implicitly minimize a squared loss. In addition, we analyze universality, consistency, and spectral filtering properties. Our theoretical results provide valuable insights into assessing the advantages and limitations of existing pairwise learning methods.
Semitransitive subspaces of operators
Bernik, J.; Drnovšek, R.; Hadwin, D.; Jafarian, A.; Bukovšek, D.K.; Košir, T.; Fijavž, M.K.; Laffey, T.; Livshits, L.; Mastnak, M.; Meshulam, R.; Müller, Vladimír; Nordgren, E.; Okniński, J.; Omladič, M.; Radjavi, H.; Sourour, A.; Timoney, R.
2006-01-01
Roč. 15, č. 1 (2006), s. 225-238 E-ISSN 1081-3810 Institutional research plan: CEZ:AV0Z10190503 Keywords : semitransitive subspaces Subject RIV: BA - General Mathematics Impact factor: 0.322, year: 2006 http://www.math.technion.ac.il/iic/ ela
A Wavelet Kernel-Based Primal Twin Support Vector Machine for Economic Development Prediction
Fang Su
2013-01-01
Full Text Available Economic development forecasting allows planners to choose the right strategies for the future. This study is to propose economic development prediction method based on the wavelet kernel-based primal twin support vector machine algorithm. As gross domestic product (GDP is an important indicator to measure economic development, economic development prediction means GDP prediction in this study. The wavelet kernel-based primal twin support vector machine algorithm can solve two smaller sized quadratic programming problems instead of solving a large one as in the traditional support vector machine algorithm. Economic development data of Anhui province from 1992 to 2009 are used to study the prediction performance of the wavelet kernel-based primal twin support vector machine algorithm. The comparison of mean error of economic development prediction between wavelet kernel-based primal twin support vector machine and traditional support vector machine models trained by the training samples with the 3–5 dimensional input vectors, respectively, is given in this paper. The testing results show that the economic development prediction accuracy of the wavelet kernel-based primal twin support vector machine model is better than that of traditional support vector machine.
Kernel method for clustering based on optimal target vector
Angelini, Leonardo; Marinazzo, Daniele; Pellicoro, Mario; Stramaglia, Sebastiano
2006-01-01
We introduce Ising models, suitable for dichotomic clustering, with couplings that are (i) both ferro- and anti-ferromagnetic (ii) depending on the whole data-set and not only on pairs of samples. Couplings are determined exploiting the notion of optimal target vector, here introduced, a link between kernel supervised and unsupervised learning. The effectiveness of the method is shown in the case of the well-known iris data-set and in benchmarks of gene expression levels, where it works better than existing methods for dichotomic clustering
Epileptic Seizure Detection with Log-Euclidean Gaussian Kernel-Based Sparse Representation.
Yuan, Shasha; Zhou, Weidong; Wu, Qi; Zhang, Yanli
2016-05-01
Epileptic seizure detection plays an important role in the diagnosis of epilepsy and reducing the massive workload of reviewing electroencephalography (EEG) recordings. In this work, a novel algorithm is developed to detect seizures employing log-Euclidean Gaussian kernel-based sparse representation (SR) in long-term EEG recordings. Unlike the traditional SR for vector data in Euclidean space, the log-Euclidean Gaussian kernel-based SR framework is proposed for seizure detection in the space of the symmetric positive definite (SPD) matrices, which form a Riemannian manifold. Since the Riemannian manifold is nonlinear, the log-Euclidean Gaussian kernel function is applied to embed it into a reproducing kernel Hilbert space (RKHS) for performing SR. The EEG signals of all channels are divided into epochs and the SPD matrices representing EEG epochs are generated by covariance descriptors. Then, the testing samples are sparsely coded over the dictionary composed by training samples utilizing log-Euclidean Gaussian kernel-based SR. The classification of testing samples is achieved by computing the minimal reconstructed residuals. The proposed method is evaluated on the Freiburg EEG dataset of 21 patients and shows its notable performance on both epoch-based and event-based assessments. Moreover, this method handles multiple channels of EEG recordings synchronously which is more speedy and efficient than traditional seizure detection methods.
Zhong, Shangping; Chen, Tianshun; He, Fengying; Niu, Yuzhen
2014-09-01
For a practical pattern classification task solved by kernel methods, the computing time is mainly spent on kernel learning (or training). However, the current kernel learning approaches are based on local optimization techniques, and hard to have good time performances, especially for large datasets. Thus the existing algorithms cannot be easily extended to large-scale tasks. In this paper, we present a fast Gaussian kernel learning method by solving a specially structured global optimization (SSGO) problem. We optimize the Gaussian kernel function by using the formulated kernel target alignment criterion, which is a difference of increasing (d.i.) functions. Through using a power-transformation based convexification method, the objective criterion can be represented as a difference of convex (d.c.) functions with a fixed power-transformation parameter. And the objective programming problem can then be converted to a SSGO problem: globally minimizing a concave function over a convex set. The SSGO problem is classical and has good solvability. Thus, to find the global optimal solution efficiently, we can adopt the improved Hoffman's outer approximation method, which need not repeat the searching procedure with different starting points to locate the best local minimum. Also, the proposed method can be proven to converge to the global solution for any classification task. We evaluate the proposed method on twenty benchmark datasets, and compare it with four other Gaussian kernel learning methods. Experimental results show that the proposed method stably achieves both good time-efficiency performance and good classification performance. Copyright © 2014 Elsevier Ltd. All rights reserved.
Multiple kernel SVR based on the MRE for remote sensing water depth fusion detection
Wang, Jinjin; Ma, Yi; Zhang, Jingyu
2018-03-01
Remote sensing has an important means of water depth detection in coastal shallow waters and reefs. Support vector regression (SVR) is a machine learning method which is widely used in data regression. In this paper, SVR is used to remote sensing multispectral bathymetry. Aiming at the problem that the single-kernel SVR method has a large error in shallow water depth inversion, the mean relative error (MRE) of different water depth is retrieved as a decision fusion factor with single kernel SVR method, a multi kernel SVR fusion method based on the MRE is put forward. And taking the North Island of the Xisha Islands in China as an experimentation area, the comparison experiments with the single kernel SVR method and the traditional multi-bands bathymetric method are carried out. The results show that: 1) In range of 0 to 25 meters, the mean absolute error(MAE)of the multi kernel SVR fusion method is 1.5m,the MRE is 13.2%; 2) Compared to the 4 single kernel SVR method, the MRE of the fusion method reduced 1.2% (1.9%) 3.4% (1.8%), and compared to traditional multi-bands method, the MRE reduced 1.9%; 3) In 0-5m depth section, compared to the single kernel method and the multi-bands method, the MRE of fusion method reduced 13.5% to 44.4%, and the distribution of points is more concentrated relative to y=x.
Probabilistic wind power forecasting based on logarithmic transformation and boundary kernel
Zhang, Yao; Wang, Jianxue; Luo, Xu
2015-01-01
Highlights: • Quantitative information on the uncertainty of wind power generation. • Kernel density estimator provides non-Gaussian predictive distributions. • Logarithmic transformation reduces the skewness of wind power density. • Boundary kernel method eliminates the density leakage near the boundary. - Abstracts: Probabilistic wind power forecasting not only produces the expectation of wind power output, but also gives quantitative information on the associated uncertainty, which is essential for making better decisions about power system and market operations with the increasing penetration of wind power generation. This paper presents a novel kernel density estimator for probabilistic wind power forecasting, addressing two characteristics of wind power which have adverse impacts on the forecast accuracy, namely, the heavily skewed and double-bounded nature of wind power density. Logarithmic transformation is used to reduce the skewness of wind power density, which improves the effectiveness of the kernel density estimator in a transformed scale. Transformations partially relieve the boundary effect problem of the kernel density estimator caused by the double-bounded nature of wind power density. However, the case study shows that there are still some serious problems of density leakage after the transformation. In order to solve this problem in the transformed scale, a boundary kernel method is employed to eliminate the density leak at the bounds of wind power distribution. The improvement of the proposed method over the standard kernel density estimator is demonstrated by short-term probabilistic forecasting results based on the data from an actual wind farm. Then, a detailed comparison is carried out of the proposed method and some existing probabilistic forecasting methods
Kernel-based whole-genome prediction of complex traits: a review.
Morota, Gota; Gianola, Daniel
2014-01-01
Prediction of genetic values has been a focus of applied quantitative genetics since the beginning of the 20th century, with renewed interest following the advent of the era of whole genome-enabled prediction. Opportunities offered by the emergence of high-dimensional genomic data fueled by post-Sanger sequencing technologies, especially molecular markers, have driven researchers to extend Ronald Fisher and Sewall Wright's models to confront new challenges. In particular, kernel methods are gaining consideration as a regression method of choice for genome-enabled prediction. Complex traits are presumably influenced by many genomic regions working in concert with others (clearly so when considering pathways), thus generating interactions. Motivated by this view, a growing number of statistical approaches based on kernels attempt to capture non-additive effects, either parametrically or non-parametrically. This review centers on whole-genome regression using kernel methods applied to a wide range of quantitative traits of agricultural importance in animals and plants. We discuss various kernel-based approaches tailored to capturing total genetic variation, with the aim of arriving at an enhanced predictive performance in the light of available genome annotation information. Connections between prediction machines born in animal breeding, statistics, and machine learning are revisited, and their empirical prediction performance is discussed. Overall, while some encouraging results have been obtained with non-parametric kernels, recovering non-additive genetic variation in a validation dataset remains a challenge in quantitative genetics.
Kernel-based whole-genome prediction of complex traits: a review
Gota eMorota
2014-10-01
Full Text Available Prediction of genetic values has been a focus of applied quantitative genetics since the beginning of the 20th century, with renewed interest following the advent of the era of whole genome-enabled prediction. Opportunities offered by the emergence of high-dimensional genomic data fueled by post-Sanger sequencing technologies, especially molecular markers, have driven researchers to extend Ronald Fisher and Sewall Wright's models to confront new challenges. In particular, kernel methods are gaining consideration as a regression method of choice for genome-enabled prediction. Complex traits are presumably influenced by many genomic regions working in concert with others (clearly so when considering pathways, thus generating interactions. Motivated by this view, a growing number of statistical approaches based on kernels attempt to capture non-additive effects, either parametrically or non-parametrically. This review centers on whole-genome regression using kernel methods applied to a wide range of quantitative traits of agricultural importance in animals and plants. We discuss various kernel-based approaches tailored to capturing total genetic variation, with the aim of arriving at an enhanced predictive performance in the light of available genome annotation information. Connections between prediction machines born in animal breeding, statistics, and machine learning are revisited, and their empirical prediction performance is discussed. Overall, while some encouraging results have been obtained with non-parametric kernels, recovering non-additive genetic variation in a validation dataset remains a challenge in quantitative genetics.
Feature Selection and Kernel Learning for Local Learning-Based Clustering.
Zeng, Hong; Cheung, Yiu-ming
2011-08-01
The performance of the most clustering algorithms highly relies on the representation of data in the input space or the Hilbert space of kernel methods. This paper is to obtain an appropriate data representation through feature selection or kernel learning within the framework of the Local Learning-Based Clustering (LLC) (Wu and Schölkopf 2006) method, which can outperform the global learning-based ones when dealing with the high-dimensional data lying on manifold. Specifically, we associate a weight to each feature or kernel and incorporate it into the built-in regularization of the LLC algorithm to take into account the relevance of each feature or kernel for the clustering. Accordingly, the weights are estimated iteratively in the clustering process. We show that the resulting weighted regularization with an additional constraint on the weights is equivalent to a known sparse-promoting penalty. Hence, the weights of those irrelevant features or kernels can be shrunk toward zero. Extensive experiments show the efficacy of the proposed methods on the benchmark data sets.
Active Subspaces for Wind Plant Surrogate Modeling
King, Ryan N [National Renewable Energy Laboratory (NREL), Golden, CO (United States); Quick, Julian [National Renewable Energy Laboratory (NREL), Golden, CO (United States); Dykes, Katherine L [National Renewable Energy Laboratory (NREL), Golden, CO (United States); Adcock, Christiane [Massachusetts Institute of Technology
2018-01-12
Understanding the uncertainty in wind plant performance is crucial to their cost-effective design and operation. However, conventional approaches to uncertainty quantification (UQ), such as Monte Carlo techniques or surrogate modeling, are often computationally intractable for utility-scale wind plants because of poor congergence rates or the curse of dimensionality. In this paper we demonstrate that wind plant power uncertainty can be well represented with a low-dimensional active subspace, thereby achieving a significant reduction in the dimension of the surrogate modeling problem. We apply the active sub-spaces technique to UQ of plant power output with respect to uncertainty in turbine axial induction factors, and find a single active subspace direction dominates the sensitivity in power output. When this single active subspace direction is used to construct a quadratic surrogate model, the number of model unknowns can be reduced by up to 3 orders of magnitude without compromising performance on unseen test data. We conclude that the dimension reduction achieved with active subspaces makes surrogate-based UQ approaches tractable for utility-scale wind plants.
Jose M. Bernal-de-Lázaro
2016-05-01
Full Text Available This article summarizes the main contributions of the PhD thesis titled: "Application of learning techniques based on kernel methods for the fault diagnosis in Industrial processes". This thesis focuses on the analysis and design of fault diagnosis systems (DDF based on historical data. Specifically this thesis provides: (1 new criteria for adjustment of the kernel methods used to select features with a high discriminative capacity for the fault diagnosis tasks, (2 a proposed approach process monitoring using statistical techniques multivariate that incorporates a reinforced information concerning to the dynamics of the Hotelling's T2 and SPE statistics, whose combination with kernel methods improves the detection of small-magnitude faults; (3 an robustness index to compare the diagnosis classifiers performance taking into account their insensitivity to possible noise and disturbance on historical data.
Sorption Kinetics for the Removal of Cadmium and Zinc onto Palm Kernel Shell Based Activated Carbon
Muhammad Muhammad
2010-12-01
Full Text Available The kinetics and mechanism of cadmium and zinc adsorption on palm kernel shell based activated carbons (PKSAC have been studied. A series of batch laboratory studies were conducted in order to investigate the suitability of palm kernel shell based activated carbon (PKSAC for the removal of cadmium (cadmium ions and zinc (zinc ions from their aqueous solutions. All batch experiments were carried out at pH 7.0 and a constant temperature of 30+-1°C using an incubator shaker that operated at 150 rpm. The kinetics investigated includes the pseudo first order, the pseudo-second order and the intraparticle diffusion models. The pseudo-second order model correlate excellently the experimental data, suggesting that chemisorption processes could be the rate-limiting step. Keywords: adsorption, cadmium, kinetics, palm kernel shell, zinc
Construction of phylogenetic trees by kernel-based comparative analysis of metabolic networks.
Oh, S June; Joung, Je-Gun; Chang, Jeong-Ho; Zhang, Byoung-Tak
2006-06-06
To infer the tree of life requires knowledge of the common characteristics of each species descended from a common ancestor as the measuring criteria and a method to calculate the distance between the resulting values of each measure. Conventional phylogenetic analysis based on genomic sequences provides information about the genetic relationships between different organisms. In contrast, comparative analysis of metabolic pathways in different organisms can yield insights into their functional relationships under different physiological conditions. However, evaluating the similarities or differences between metabolic networks is a computationally challenging problem, and systematic methods of doing this are desirable. Here we introduce a graph-kernel method for computing the similarity between metabolic networks in polynomial time, and use it to profile metabolic pathways and to construct phylogenetic trees. To compare the structures of metabolic networks in organisms, we adopted the exponential graph kernel, which is a kernel-based approach with a labeled graph that includes a label matrix and an adjacency matrix. To construct the phylogenetic trees, we used an unweighted pair-group method with arithmetic mean, i.e., a hierarchical clustering algorithm. We applied the kernel-based network profiling method in a comparative analysis of nine carbohydrate metabolic networks from 81 biological species encompassing Archaea, Eukaryota, and Eubacteria. The resulting phylogenetic hierarchies generally support the tripartite scheme of three domains rather than the two domains of prokaryotes and eukaryotes. By combining the kernel machines with metabolic information, the method infers the context of biosphere development that covers physiological events required for adaptation by genetic reconstruction. The results show that one may obtain a global view of the tree of life by comparing the metabolic pathway structures using meta-level information rather than sequence
Construction of phylogenetic trees by kernel-based comparative analysis of metabolic networks
Chang Jeong-Ho
2006-06-01
Full Text Available Abstract Background To infer the tree of life requires knowledge of the common characteristics of each species descended from a common ancestor as the measuring criteria and a method to calculate the distance between the resulting values of each measure. Conventional phylogenetic analysis based on genomic sequences provides information about the genetic relationships between different organisms. In contrast, comparative analysis of metabolic pathways in different organisms can yield insights into their functional relationships under different physiological conditions. However, evaluating the similarities or differences between metabolic networks is a computationally challenging problem, and systematic methods of doing this are desirable. Here we introduce a graph-kernel method for computing the similarity between metabolic networks in polynomial time, and use it to profile metabolic pathways and to construct phylogenetic trees. Results To compare the structures of metabolic networks in organisms, we adopted the exponential graph kernel, which is a kernel-based approach with a labeled graph that includes a label matrix and an adjacency matrix. To construct the phylogenetic trees, we used an unweighted pair-group method with arithmetic mean, i.e., a hierarchical clustering algorithm. We applied the kernel-based network profiling method in a comparative analysis of nine carbohydrate metabolic networks from 81 biological species encompassing Archaea, Eukaryota, and Eubacteria. The resulting phylogenetic hierarchies generally support the tripartite scheme of three domains rather than the two domains of prokaryotes and eukaryotes. Conclusion By combining the kernel machines with metabolic information, the method infers the context of biosphere development that covers physiological events required for adaptation by genetic reconstruction. The results show that one may obtain a global view of the tree of life by comparing the metabolic pathway
Seismic noise attenuation using an online subspace tracking algorithm
Zhou, Yatong; Li, Shuhua; Zhang, D.; Chen, Yangkang
2018-01-01
We propose a new low-rank based noise attenuation method using an efficient algorithm for tracking subspaces from highly corrupted seismic observations. The subspace tracking algorithm requires only basic linear algebraic manipulations. The algorithm is derived by analysing incremental gradient
Kernel Method Based Human Model for Enhancing Interactive Evolutionary Optimization
Zhao, Qiangfu; Liu, Yong
2015-01-01
A fitness landscape presents the relationship between individual and its reproductive success in evolutionary computation (EC). However, discrete and approximate landscape in an original search space may not support enough and accurate information for EC search, especially in interactive EC (IEC). The fitness landscape of human subjective evaluation in IEC is very difficult and impossible to model, even with a hypothesis of what its definition might be. In this paper, we propose a method to establish a human model in projected high dimensional search space by kernel classification for enhancing IEC search. Because bivalent logic is a simplest perceptual paradigm, the human model is established by considering this paradigm principle. In feature space, we design a linear classifier as a human model to obtain user preference knowledge, which cannot be supported linearly in original discrete search space. The human model is established by this method for predicting potential perceptual knowledge of human. With the human model, we design an evolution control method to enhance IEC search. From experimental evaluation results with a pseudo-IEC user, our proposed model and method can enhance IEC search significantly. PMID:25879050
Suetani, Hiromichi; Iba, Yukito; Aihara, Kazuyuki
2006-01-01
An approach based on the kernel methods for capturing the nonlinear interdependence between two signals is introduced. It is demonstrated that the proposed approach is useful for characterizing generalized synchronization with a successful simple example. An attempt to choose an optimal kernel parameter based on cross validation is also discussed. (author)
Combined Kernel-Based BDT-SMO Classification of Hyperspectral Fused Images
Fenghua Huang
2014-01-01
Full Text Available To solve the poor generalization and flexibility problems that single kernel SVM classifiers have while classifying combined spectral and spatial features, this paper proposed a solution to improve the classification accuracy and efficiency of hyperspectral fused images: (1 different radial basis kernel functions (RBFs are employed for spectral and textural features, and a new combined radial basis kernel function (CRBF is proposed by combining them in a weighted manner; (2 the binary decision tree-based multiclass SMO (BDT-SMO is used in the classification of hyperspectral fused images; (3 experiments are carried out, where the single radial basis function- (SRBF- based BDT-SMO classifier and the CRBF-based BDT-SMO classifier are used, respectively, to classify the land usages of hyperspectral fused images, and genetic algorithms (GA are used to optimize the kernel parameters of the classifiers. The results show that, compared with SRBF, CRBF-based BDT-SMO classifiers display greater classification accuracy and efficiency.
Chen, Lili; Zhang, Xi; Wang, Hui
2015-05-01
Obstructive sleep apnea (OSA) is a common sleep disorder that often remains undiagnosed, leading to an increased risk of developing cardiovascular diseases. Polysomnogram (PSG) is currently used as a golden standard for screening OSA. However, because it is time consuming, expensive and causes discomfort, alternative techniques based on a reduced set of physiological signals are proposed to solve this problem. This study proposes a convenient non-parametric kernel density-based approach for detection of OSA using single-lead electrocardiogram (ECG) recordings. Selected physiologically interpretable features are extracted from segmented RR intervals, which are obtained from ECG signals. These features are fed into the kernel density classifier to detect apnea event and bandwidths for density of each class (normal or apnea) are automatically chosen through an iterative bandwidth selection algorithm. To validate the proposed approach, RR intervals are extracted from ECG signals of 35 subjects obtained from a sleep apnea database ( http://physionet.org/cgi-bin/atm/ATM ). The results indicate that the kernel density classifier, with two features for apnea event detection, achieves a mean accuracy of 82.07 %, with mean sensitivity of 83.23 % and mean specificity of 80.24 %. Compared with other existing methods, the proposed kernel density approach achieves a comparably good performance but by using fewer features without significantly losing discriminant power, which indicates that it could be widely used for home-based screening or diagnosis of OSA.
A Novel Mittag-Leffler Kernel Based Hybrid Fault Diagnosis Method for Wheeled Robot Driving System
Xianfeng Yuan
2015-01-01
presents a novel hybrid fault diagnosis framework based on Mittag-Leffler kernel (ML-kernel support vector machine (SVM and Dempster-Shafer (D-S fusion. Using sensor data sampled under different running conditions, the proposed approach initially establishes multiple principal component analysis (PCA models for fault feature extraction. The fault feature vectors are then applied to train the probabilistic SVM (PSVM classifiers that arrive at a preliminary fault diagnosis. To improve the accuracy of preliminary results, a novel ML-kernel based PSVM classifier is proposed in this paper, and the positive definiteness of the ML-kernel is proved as well. The basic probability assignments (BPAs are defined based on the preliminary fault diagnosis results and their confidence values. Eventually, the final fault diagnosis result is archived by the fusion of the BPAs. Experimental results show that the proposed framework not only is capable of detecting and identifying the faults in the robot driving system, but also has better performance in stability and diagnosis accuracy compared with the traditional methods.
On the dimension of subspaces with bounded Schmidt rank
Cubitt, Toby; Montanaro, Ashley; Winter, Andreas
2008-01-01
We consider the question of how large a subspace of a given bipartite quantum system can be when the subspace contains only highly entangled states. This is motivated in part by results of Hayden et al. [e-print arXiv:quant-ph/0407049; Commun. Math. Phys., 265, 95 (2006)], which show that in large dxd-dimensional systems there exist random subspaces of dimension almost d 2 , all of whose states have entropy of entanglement at least log d-O(1). It is also a generalization of results on the dimension of completely entangled subspaces, which have connections with the construction of unextendible product bases. Here we take as entanglement measure the Schmidt rank, and determine, for every pair of local dimensions d A and d B , and every r, the largest dimension of a subspace consisting only of entangled states of Schmidt rank r or larger. This exact answer is a significant improvement on the best bounds that can be obtained using the random subspace techniques in Hayden et al. We also determine the converse: the largest dimension of a subspace with an upper bound on the Schmidt rank. Finally, we discuss the question of subspaces containing only states with Schmidt equal to r
Uchida, Isao; Yamada, Yasuhiko; Yamashita, Takashi; Okigaki, Shigeyasu; Oyamada, Hiyoshimaru; Ito, Akira.
1995-01-01
In radiotherapy with radiopharmaceuticals, more accurate estimates of the three-dimensional (3-D) distribution of absorbed dose is important in specifying the activity to be administered to patients to deliver a prescribed absorbed dose to target volumes without exceeding the toxicity limit of normal tissues in the body. A calculation algorithm for the purpose has already been developed by the authors. An accurate 3-D distribution of absorbed dose based on the algorithm is given by convolution of the 3-D dose matrix for a unit cubic voxel containing unit cumulated activity, which is obtained by transforming a dose point kernel into a 3-D cubic dose matrix, with the 3-D cumulated activity distribution given by the same voxel size. However, beta-dose point kernels affecting accurate estimates of the 3-D absorbed dose distribution have been different among the investigators. The purpose of this study is to elucidate how different beta-dose point kernels in water influence on the estimates of the absorbed dose distribution due to the dose point kernel convolution method by the authors. Computer simulations were performed using the MIRD thyroid and lung phantoms under assumption of uniform activity distribution of 32 P. Using beta-dose point kernels derived from Monte Carlo simulations (EGS-4 or ACCEPT computer code), the differences among their point kernels gave little differences for the mean and maximum absorbed dose estimates for the MIRD phantoms used. In the estimates of mean and maximum absorbed doses calculated using different cubic voxel sizes (4x4x4 mm and 8x8x8 mm) for the MIRD thyroid phantom, the maximum absorbed doses for the 4x4x4 mm-voxel were estimated approximately 7% greater than the cases of the 8x8x8 mm-voxel. They were found in every beta-dose point kernel used in this study. On the other hand, the percentage difference of the mean absorbed doses in the both voxel sizes for each beta-dose point kernel was less than approximately 0.6%. (author)
Jiang, Qingchao; Yan, Xuefeng; Lv, Zhaomin; Guo, Meijin [East China University of Science and Technology, Shanghai (China)
2013-06-15
Considering that kernel entropy component analysis (KECA) is a promising new method of nonlinear data transformation and dimensionality reduction, a KECA based method is proposed for nonlinear chemical process monitoring. In this method, an angle-based statistic is designed because KECA reveals structure related to the Renyi entropy of input space data set, and the transformed data sets are produced with a distinct angle-based structure. Based on the angle difference between normal status and current sample data, the current status can be monitored effectively. And, the confidence limit of the angle-based statistics is determined by kernel density estimation based on sample data of the normal status. The effectiveness of the proposed method is demonstrated by case studies on both a numerical process and a simulated continuous stirred tank reactor (CSTR) process. The KECA based method can be an effective method for nonlinear chemical process monitoring.
Jiang, Qingchao; Yan, Xuefeng; Lv, Zhaomin; Guo, Meijin
2013-01-01
Considering that kernel entropy component analysis (KECA) is a promising new method of nonlinear data transformation and dimensionality reduction, a KECA based method is proposed for nonlinear chemical process monitoring. In this method, an angle-based statistic is designed because KECA reveals structure related to the Renyi entropy of input space data set, and the transformed data sets are produced with a distinct angle-based structure. Based on the angle difference between normal status and current sample data, the current status can be monitored effectively. And, the confidence limit of the angle-based statistics is determined by kernel density estimation based on sample data of the normal status. The effectiveness of the proposed method is demonstrated by case studies on both a numerical process and a simulated continuous stirred tank reactor (CSTR) process. The KECA based method can be an effective method for nonlinear chemical process monitoring
A Precise Drunk Driving Detection Using Weighted Kernel Based on Electrocardiogram.
Wu, Chung Kit; Tsang, Kim Fung; Chi, Hao Ran; Hung, Faan Hei
2016-05-09
Globally, 1.2 million people die and 50 million people are injured annually due to traffic accidents. These traffic accidents cost $500 billion dollars. Drunk drivers are found in 40% of the traffic crashes. Existing drunk driving detection (DDD) systems do not provide accurate detection and pre-warning concurrently. Electrocardiogram (ECG) is a proven biosignal that accurately and simultaneously reflects human's biological status. In this letter, a classifier for DDD based on ECG is investigated in an attempt to reduce traffic accidents caused by drunk drivers. At this point, it appears that there is no known research or literature found on ECG classifier for DDD. To identify drunk syndromes, the ECG signals from drunk drivers are studied and analyzed. As such, a precise ECG-based DDD (ECG-DDD) using a weighted kernel is developed. From the measurements, 10 key features of ECG signals were identified. To incorporate the important features, the feature vectors are weighted in the customization of kernel functions. Four commonly adopted kernel functions are studied. Results reveal that weighted feature vectors improve the accuracy by 11% compared to the computation using the prime kernel. Evaluation shows that ECG-DDD improved the accuracy by 8% to 18% compared to prevailing methods.
A Precise Drunk Driving Detection Using Weighted Kernel Based on Electrocardiogram
Chung Kit Wu
2016-05-01
Full Text Available Globally, 1.2 million people die and 50 million people are injured annually due to traffic accidents. These traffic accidents cost $500 billion dollars. Drunk drivers are found in 40% of the traffic crashes. Existing drunk driving detection (DDD systems do not provide accurate detection and pre-warning concurrently. Electrocardiogram (ECG is a proven biosignal that accurately and simultaneously reflects human’s biological status. In this letter, a classifier for DDD based on ECG is investigated in an attempt to reduce traffic accidents caused by drunk drivers. At this point, it appears that there is no known research or literature found on ECG classifier for DDD. To identify drunk syndromes, the ECG signals from drunk drivers are studied and analyzed. As such, a precise ECG-based DDD (ECG-DDD using a weighted kernel is developed. From the measurements, 10 key features of ECG signals were identified. To incorporate the important features, the feature vectors are weighted in the customization of kernel functions. Four commonly adopted kernel functions are studied. Results reveal that weighted feature vectors improve the accuracy by 11% compared to the computation using the prime kernel. Evaluation shows that ECG-DDD improved the accuracy by 8% to 18% compared to prevailing methods.
Chen Qiang; Ren Xuemei; Na Jing
2011-01-01
Highlights: Model uncertainty of the system is approximated by multiple-kernel LSSVM. Approximation errors and disturbances are compensated in the controller design. Asymptotical anti-synchronization is achieved with model uncertainty and disturbances. Abstract: In this paper, we propose a robust anti-synchronization scheme based on multiple-kernel least squares support vector machine (MK-LSSVM) modeling for two uncertain chaotic systems. The multiple-kernel regression, which is a linear combination of basic kernels, is designed to approximate system uncertainties by constructing a multiple-kernel Lagrangian function and computing the corresponding regression parameters. Then, a robust feedback control based on MK-LSSVM modeling is presented and an improved update law is employed to estimate the unknown bound of the approximation error. The proposed control scheme can guarantee the asymptotic convergence of the anti-synchronization errors in the presence of system uncertainties and external disturbances. Numerical examples are provided to show the effectiveness of the proposed method.
Semi-supervised weighted kernel clustering based on gravitational search for fault diagnosis.
Li, Chaoshun; Zhou, Jianzhong
2014-09-01
Supervised learning method, like support vector machine (SVM), has been widely applied in diagnosing known faults, however this kind of method fails to work correctly when new or unknown fault occurs. Traditional unsupervised kernel clustering can be used for unknown fault diagnosis, but it could not make use of the historical classification information to improve diagnosis accuracy. In this paper, a semi-supervised kernel clustering model is designed to diagnose known and unknown faults. At first, a novel semi-supervised weighted kernel clustering algorithm based on gravitational search (SWKC-GS) is proposed for clustering of dataset composed of labeled and unlabeled fault samples. The clustering model of SWKC-GS is defined based on wrong classification rate of labeled samples and fuzzy clustering index on the whole dataset. Gravitational search algorithm (GSA) is used to solve the clustering model, while centers of clusters, feature weights and parameter of kernel function are selected as optimization variables. And then, new fault samples are identified and diagnosed by calculating the weighted kernel distance between them and the fault cluster centers. If the fault samples are unknown, they will be added in historical dataset and the SWKC-GS is used to partition the mixed dataset and update the clustering results for diagnosing new fault. In experiments, the proposed method has been applied in fault diagnosis for rotatory bearing, while SWKC-GS has been compared not only with traditional clustering methods, but also with SVM and neural network, for known fault diagnosis. In addition, the proposed method has also been applied in unknown fault diagnosis. The results have shown effectiveness of the proposed method in achieving expected diagnosis accuracy for both known and unknown faults of rotatory bearing. Copyright © 2014 ISA. Published by Elsevier Ltd. All rights reserved.
Central subspace dimensionality reduction using covariance operators.
Kim, Minyoung; Pavlovic, Vladimir
2011-04-01
We consider the task of dimensionality reduction informed by real-valued multivariate labels. The problem is often treated as Dimensionality Reduction for Regression (DRR), whose goal is to find a low-dimensional representation, the central subspace, of the input data that preserves the statistical correlation with the targets. A class of DRR methods exploits the notion of inverse regression (IR) to discover central subspaces. Whereas most existing IR techniques rely on explicit output space slicing, we propose a novel method called the Covariance Operator Inverse Regression (COIR) that generalizes IR to nonlinear input/output spaces without explicit target slicing. COIR's unique properties make DRR applicable to problem domains with high-dimensional output data corrupted by potentially significant amounts of noise. Unlike recent kernel dimensionality reduction methods that employ iterative nonconvex optimization, COIR yields a closed-form solution. We also establish the link between COIR, other DRR techniques, and popular supervised dimensionality reduction methods, including canonical correlation analysis and linear discriminant analysis. We then extend COIR to semi-supervised settings where many of the input points lack their labels. We demonstrate the benefits of COIR on several important regression problems in both fully supervised and semi-supervised settings.
Estimation of the applicability domain of kernel-based machine learning models for virtual screening
Fechner Nikolas
2010-03-01
Full Text Available Abstract Background The virtual screening of large compound databases is an important application of structural-activity relationship models. Due to the high structural diversity of these data sets, it is impossible for machine learning based QSAR models, which rely on a specific training set, to give reliable results for all compounds. Thus, it is important to consider the subset of the chemical space in which the model is applicable. The approaches to this problem that have been published so far mostly use vectorial descriptor representations to define this domain of applicability of the model. Unfortunately, these cannot be extended easily to structured kernel-based machine learning models. For this reason, we propose three approaches to estimate the domain of applicability of a kernel-based QSAR model. Results We evaluated three kernel-based applicability domain estimations using three different structured kernels on three virtual screening tasks. Each experiment consisted of the training of a kernel-based QSAR model using support vector regression and the ranking of a disjoint screening data set according to the predicted activity. For each prediction, the applicability of the model for the respective compound is quantitatively described using a score obtained by an applicability domain formulation. The suitability of the applicability domain estimation is evaluated by comparing the model performance on the subsets of the screening data sets obtained by different thresholds for the applicability scores. This comparison indicates that it is possible to separate the part of the chemspace, in which the model gives reliable predictions, from the part consisting of structures too dissimilar to the training set to apply the model successfully. A closer inspection reveals that the virtual screening performance of the model is considerably improved if half of the molecules, those with the lowest applicability scores, are omitted from the screening
Fechner, Nikolas; Jahn, Andreas; Hinselmann, Georg; Zell, Andreas
2010-03-11
The virtual screening of large compound databases is an important application of structural-activity relationship models. Due to the high structural diversity of these data sets, it is impossible for machine learning based QSAR models, which rely on a specific training set, to give reliable results for all compounds. Thus, it is important to consider the subset of the chemical space in which the model is applicable. The approaches to this problem that have been published so far mostly use vectorial descriptor representations to define this domain of applicability of the model. Unfortunately, these cannot be extended easily to structured kernel-based machine learning models. For this reason, we propose three approaches to estimate the domain of applicability of a kernel-based QSAR model. We evaluated three kernel-based applicability domain estimations using three different structured kernels on three virtual screening tasks. Each experiment consisted of the training of a kernel-based QSAR model using support vector regression and the ranking of a disjoint screening data set according to the predicted activity. For each prediction, the applicability of the model for the respective compound is quantitatively described using a score obtained by an applicability domain formulation. The suitability of the applicability domain estimation is evaluated by comparing the model performance on the subsets of the screening data sets obtained by different thresholds for the applicability scores. This comparison indicates that it is possible to separate the part of the chemspace, in which the model gives reliable predictions, from the part consisting of structures too dissimilar to the training set to apply the model successfully. A closer inspection reveals that the virtual screening performance of the model is considerably improved if half of the molecules, those with the lowest applicability scores, are omitted from the screening. The proposed applicability domain formulations
Yang Zhao
Full Text Available GWAS has facilitated greatly the discovery of risk SNPs associated with complex diseases. Traditional methods analyze SNP individually and are limited by low power and reproducibility since correction for multiple comparisons is necessary. Several methods have been proposed based on grouping SNPs into SNP sets using biological knowledge and/or genomic features. In this article, we compare the linear kernel machine based test (LKM and principal components analysis based approach (PCA using simulated datasets under the scenarios of 0 to 3 causal SNPs, as well as simple and complex linkage disequilibrium (LD structures of the simulated regions. Our simulation study demonstrates that both LKM and PCA can control the type I error at the significance level of 0.05. If the causal SNP is in strong LD with the genotyped SNPs, both the PCA with a small number of principal components (PCs and the LKM with kernel of linear or identical-by-state function are valid tests. However, if the LD structure is complex, such as several LD blocks in the SNP set, or when the causal SNP is not in the LD block in which most of the genotyped SNPs reside, more PCs should be included to capture the information of the causal SNP. Simulation studies also demonstrate the ability of LKM and PCA to combine information from multiple causal SNPs and to provide increased power over individual SNP analysis. We also apply LKM and PCA to analyze two SNP sets extracted from an actual GWAS dataset on non-small cell lung cancer.
Grey Language Hesitant Fuzzy Group Decision Making Method Based on Kernel and Grey Scale.
Li, Qingsheng; Diao, Yuzhu; Gong, Zaiwu; Hu, Aqin
2018-03-02
Based on grey language multi-attribute group decision making, a kernel and grey scale scoring function is put forward according to the definition of grey language and the meaning of the kernel and grey scale. The function introduces grey scale into the decision-making method to avoid information distortion. This method is applied to the grey language hesitant fuzzy group decision making, and the grey correlation degree is used to sort the schemes. The effectiveness and practicability of the decision-making method are further verified by the industry chain sustainable development ability evaluation example of a circular economy. Moreover, its simplicity and feasibility are verified by comparing it with the traditional grey language decision-making method and the grey language hesitant fuzzy weighted arithmetic averaging (GLHWAA) operator integration method after determining the index weight based on the grey correlation.
Pham, Binh Thai; Prakash, Indra; Tien Bui, Dieu
2018-02-01
A hybrid machine learning approach of Random Subspace (RSS) and Classification And Regression Trees (CART) is proposed to develop a model named RSSCART for spatial prediction of landslides. This model is a combination of the RSS method which is known as an efficient ensemble technique and the CART which is a state of the art classifier. The Luc Yen district of Yen Bai province, a prominent landslide prone area of Viet Nam, was selected for the model development. Performance of the RSSCART model was evaluated through the Receiver Operating Characteristic (ROC) curve, statistical analysis methods, and the Chi Square test. Results were compared with other benchmark landslide models namely Support Vector Machines (SVM), single CART, Naïve Bayes Trees (NBT), and Logistic Regression (LR). In the development of model, ten important landslide affecting factors related with geomorphology, geology and geo-environment were considered namely slope angles, elevation, slope aspect, curvature, lithology, distance to faults, distance to rivers, distance to roads, and rainfall. Performance of the RSSCART model (AUC = 0.841) is the best compared with other popular landslide models namely SVM (0.835), single CART (0.822), NBT (0.821), and LR (0.723). These results indicate that performance of the RSSCART is a promising method for spatial landslide prediction.
Timmerman, Marieke E; Ceulemans, Eva; De Roover, Kim; Van Leeuwen, Karla
2013-12-01
To achieve an insightful clustering of multivariate data, we propose subspace K-means. Its central idea is to model the centroids and cluster residuals in reduced spaces, which allows for dealing with a wide range of cluster types and yields rich interpretations of the clusters. We review the existing related clustering methods, including deterministic, stochastic, and unsupervised learning approaches. To evaluate subspace K-means, we performed a comparative simulation study, in which we manipulated the overlap of subspaces, the between-cluster variance, and the error variance. The study shows that the subspace K-means algorithm is sensitive to local minima but that the problem can be reasonably dealt with by using partitions of various cluster procedures as a starting point for the algorithm. Subspace K-means performs very well in recovering the true clustering across all conditions considered and appears to be superior to its competitor methods: K-means, reduced K-means, factorial K-means, mixtures of factor analyzers (MFA), and MCLUST. The best competitor method, MFA, showed a performance similar to that of subspace K-means in easy conditions but deteriorated in more difficult ones. Using data from a study on parental behavior, we show that subspace K-means analysis provides a rich insight into the cluster characteristics, in terms of both the relative positions of the clusters (via the centroids) and the shape of the clusters (via the within-cluster residuals).
Wang, Xin; Wei, Guo; Sun, Jinwei
2013-01-01
The signal reconstruction methods based on inverse modeling for the signal reconstruction of multifunctional sensors have been widely studied in recent years. To improve the accuracy, the reconstruction methods have become more and more complicated because of the increase in the model parameters and sample points. However, there is another factor that affects the reconstruction accuracy, the position of the sample points, which has not been studied. A reasonable selection of the sample points could improve the signal reconstruction quality in at least two ways: improved accuracy with the same number of sample points or the same accuracy obtained with a smaller number of sample points. Both ways are valuable for improving the accuracy and decreasing the workload, especially for large batches of multifunctional sensors. In this paper, we propose a sample selection method based on kernel-subclustering distill groupings of the sample data and produce the representation of the data set for inverse modeling. The method calculates the distance between two data points based on the kernel-induced distance instead of the conventional distance. The kernel function is a generalization of the distance metric by mapping the data that are non-separable in the original space into homogeneous groups in the high-dimensional space. The method obtained the best results compared with the other three methods in the simulation. (paper)
Haorui Liu
2016-01-01
Full Text Available In the car control systems, it is hard to measure some key vehicle states directly and accurately when running on the road and the cost of the measurement is high as well. To address these problems, a vehicle state estimation method based on the kernel principal component analysis and the improved Elman neural network is proposed. Combining with nonlinear vehicle model of three degrees of freedom (3 DOF, longitudinal, lateral, and yaw motion, this paper applies the method to the soft sensor of the vehicle states. The simulation results of the double lane change tested by Matlab/SIMULINK cosimulation prove the KPCA-IENN algorithm (kernel principal component algorithm and improved Elman neural network to be quick and precise when tracking the vehicle states within the nonlinear area. This algorithm method can meet the software performance requirements of the vehicle states estimation in precision, tracking speed, noise suppression, and other aspects.
Personal sleep pattern visualization using sequence-based kernel self-organizing map on sound data.
Wu, Hongle; Kato, Takafumi; Yamada, Tomomi; Numao, Masayuki; Fukui, Ken-Ichi
2017-07-01
We propose a method to discover sleep patterns via clustering of sound events recorded during sleep. The proposed method extends the conventional self-organizing map algorithm by kernelization and sequence-based technologies to obtain a fine-grained map that visualizes the distribution and changes of sleep-related events. We introduced features widely applied in sound processing and popular kernel functions to the proposed method to evaluate and compare performance. The proposed method provides a new aspect of sleep monitoring because the results demonstrate that sound events can be directly correlated to an individual's sleep patterns. In addition, by visualizing the transition of cluster dynamics, sleep-related sound events were found to relate to the various stages of sleep. Therefore, these results empirically warrant future study into the assessment of personal sleep quality using sound data. Copyright © 2017 Elsevier B.V. All rights reserved.
Djebbi, Ramzi
2015-08-19
The instantaneous traveltime is able to reduce the non-linearity of full waveform inversion (FWI) that originates from the wrapping of the phase. However, the adjoint state method in this case requires a total of 5 modeling calculations to compute the gradient. Also, considering the larger modeling cost for anisotropic wavefield extrapolation and the necessity to use a line-search algorithm to estimate a step length that depends on the parameters scale, we propose to calculate the gradient based on the instantaneous traveltime sensitivity kernels. We, specifically, use the sensitivity kernels computed using dynamic ray-tracing to build the gradient. The resulting update is computed using a matrix decomposition and accordingly the computational cost is reduced. We consider a simple example where an anomaly is embedded into a constant background medium and we compute the update for the VTI wave equation parameterized using vh, η and ε.
Djebbi, Ramzi; Alkhalifah, Tariq Ali
2015-01-01
The instantaneous traveltime is able to reduce the non-linearity of full waveform inversion (FWI) that originates from the wrapping of the phase. However, the adjoint state method in this case requires a total of 5 modeling calculations to compute the gradient. Also, considering the larger modeling cost for anisotropic wavefield extrapolation and the necessity to use a line-search algorithm to estimate a step length that depends on the parameters scale, we propose to calculate the gradient based on the instantaneous traveltime sensitivity kernels. We, specifically, use the sensitivity kernels computed using dynamic ray-tracing to build the gradient. The resulting update is computed using a matrix decomposition and accordingly the computational cost is reduced. We consider a simple example where an anomaly is embedded into a constant background medium and we compute the update for the VTI wave equation parameterized using vh, η and ε.
Kernel-Based Learning for Domain-Specific Relation Extraction
Basili, Roberto; Giannone, Cristina; Del Vescovo, Chiara; Moschitti, Alessandro; Naggar, Paolo
In a specific process of business intelligence, i.e. investigation on organized crime, empirical language processing technologies can play a crucial role. The analysis of transcriptions on investigative activities, such as police interrogatories, for the recognition and storage of complex relations among people and locations is a very difficult and time consuming task, ultimately based on pools of experts. We discuss here an inductive relation extraction platform that opens the way to much cheaper and consistent workflows. The presented empirical investigation shows that accurate results, comparable to the expert teams, can be achieved, and parametrization allows to fine tune the system behavior for fitting domain-specific requirements.
Chen, Xin; Xie, Penghuan; Xiong, Yonghua; He, Yong; Wu, Min
2015-01-01
Adaptive Dynamic Programming (ADP) with critic-actor architecture is an effective way to perform online learning control. To avoid the subjectivity in the design of a neural network that serves as a critic network, kernel-based adaptive critic design (ACD) was developed recently. There are two essential issues for a static kernel-based model: how to determine proper hyperparameters in advance and how to select right samples to describe the value function. They all rely on the assessment of sa...
Research on a Novel Kernel Based Grey Prediction Model and Its Applications
Xin Ma
2016-01-01
Full Text Available The discrete grey prediction models have attracted considerable interest of research due to its effectiveness to improve the modelling accuracy of the traditional grey prediction models. The autoregressive GM(1,1 model, abbreviated as ARGM(1,1, is a novel discrete grey model which is easy to use and accurate in prediction of approximate nonhomogeneous exponential time series. However, the ARGM(1,1 is essentially a linear model; thus, its applicability is still limited. In this paper a novel kernel based ARGM(1,1 model is proposed, abbreviated as KARGM(1,1. The KARGM(1,1 has a nonlinear function which can be expressed by a kernel function using the kernel method, and its modelling procedures are presented in details. Two case studies of predicting the monthly gas well production are carried out with the real world production data. The results of KARGM(1,1 model are compared to the existing discrete univariate grey prediction models, including ARGM(1,1, NDGM(1,1,k, DGM(1,1, and NGBMOP, and it is shown that the KARGM(1,1 outperforms the other four models.
Forecasting Crude Oil Price Using EEMD and RVM with Adaptive PSO-Based Kernels
Taiyong Li
2016-12-01
Full Text Available Crude oil, as one of the most important energy sources in the world, plays a crucial role in global economic events. An accurate prediction for crude oil price is an interesting and challenging task for enterprises, governments, investors, and researchers. To cope with this issue, in this paper, we proposed a method integrating ensemble empirical mode decomposition (EEMD, adaptive particle swarm optimization (APSO, and relevance vector machine (RVM—namely, EEMD-APSO-RVM—to predict crude oil price based on the “decomposition and ensemble” framework. Specifically, the raw time series of crude oil price were firstly decomposed into several intrinsic mode functions (IMFs and one residue by EEMD. Then, RVM with combined kernels was applied to predict target value for the residue and each IMF individually. To improve the prediction performance of each component, an extended particle swarm optimization (PSO was utilized to simultaneously optimize the weights and parameters of single kernels for the combined kernel of RVM. Finally, simple addition was used to aggregate all the predicted results of components into an ensemble result as the final result. Extensive experiments were conducted on the crude oil spot price of the West Texas Intermediate (WTI to illustrate and evaluate the proposed method. The experimental results are superior to those by several state-of-the-art benchmark methods in terms of root mean squared error (RMSE, mean absolute percent error (MAPE, and directional statistic (Dstat, showing that the proposed EEMD-APSO-RVM is promising for forecasting crude oil price.
Andres M. Alvarez-Meza
2017-10-01
Full Text Available We introduce Enhanced Kernel-based Relevance Analysis (EKRA that aims to support the automatic identification of brain activity patterns using electroencephalographic recordings. EKRA is a data-driven strategy that incorporates two kernel functions to take advantage of the available joint information, associating neural responses to a given stimulus condition. Regarding this, a Centered Kernel Alignment functional is adjusted to learning the linear projection that best discriminates the input feature set, optimizing the required free parameters automatically. Our approach is carried out in two scenarios: (i feature selection by computing a relevance vector from extracted neural features to facilitating the physiological interpretation of a given brain activity task, and (ii enhanced feature selection to perform an additional transformation of relevant features aiming to improve the overall identification accuracy. Accordingly, we provide an alternative feature relevance analysis strategy that allows improving the system performance while favoring the data interpretability. For the validation purpose, EKRA is tested in two well-known tasks of brain activity: motor imagery discrimination and epileptic seizure detection. The obtained results show that the EKRA approach estimates a relevant representation space extracted from the provided supervised information, emphasizing the salient input features. As a result, our proposal outperforms the state-of-the-art methods regarding brain activity discrimination accuracy with the benefit of enhanced physiological interpretation about the task at hand.
A kernel-based multi-feature image representation for histopathology image classification
Moreno J; Caicedo J Gonzalez F
2010-01-01
This paper presents a novel strategy for building a high-dimensional feature space to represent histopathology image contents. Histogram features, related to colors, textures and edges, are combined together in a unique image representation space using kernel functions. This feature space is further enhanced by the application of latent semantic analysis, to model hidden relationships among visual patterns. All that information is included in the new image representation space. Then, support vector machine classifiers are used to assign semantic labels to images. Processing and classification algorithms operate on top of kernel functions, so that; the structure of the feature space is completely controlled using similarity measures and a dual representation. The proposed approach has shown a successful performance in a classification task using a dataset with 1,502 real histopathology images in 18 different classes. The results show that our approach for histological image classification obtains an improved average performance of 20.6% when compared to a conventional classification approach based on SVM directly applied to the original kernel.
A KERNEL-BASED MULTI-FEATURE IMAGE REPRESENTATION FOR HISTOPATHOLOGY IMAGE CLASSIFICATION
J Carlos Moreno
2010-09-01
Full Text Available This paper presents a novel strategy for building a high-dimensional feature space to represent histopathology image contents. Histogram features, related to colors, textures and edges, are combined together in a unique image representation space using kernel functions. This feature space is further enhanced by the application of Latent Semantic Analysis, to model hidden relationships among visual patterns. All that information is included in the new image representation space. Then, Support Vector Machine classifiers are used to assign semantic labels to images. Processing and classification algorithms operate on top of kernel functions, so that, the structure of the feature space is completely controlled using similarity measures and a dual representation. The proposed approach has shown a successful performance in a classification task using a dataset with 1,502 real histopathology images in 18 different classes. The results show that our approach for histological image classification obtains an improved average performance of 20.6% when compared to a conventional classification approach based on SVM directly applied to the original kernel.
Chao Huang
2014-01-01
Full Text Available Financial distress prediction plays an important role in the survival of companies. In this paper, a novel biorthogonal wavelet hybrid kernel function is constructed by combining linear kernel function with biorthogonal wavelet kernel function. Besides, a new feature weighted approach is presented based on economic value added (EVA and grey relational analysis (GRA. Considering the imbalance between financially distressed companies and normal ones, the feature weighted one-class support vector machine based on biorthogonal wavelet hybrid kernel (BWH-FWOCSVM is further put forward for financial distress prediction. The empirical study with real data from the listed companies on Growth Enterprise Market (GEM in China shows that the proposed approach has good performance.
Optimized Kernel Entropy Components.
Izquierdo-Verdiguier, Emma; Laparra, Valero; Jenssen, Robert; Gomez-Chova, Luis; Camps-Valls, Gustau
2017-06-01
This brief addresses two main issues of the standard kernel entropy component analysis (KECA) algorithm: the optimization of the kernel decomposition and the optimization of the Gaussian kernel parameter. KECA roughly reduces to a sorting of the importance of kernel eigenvectors by entropy instead of variance, as in the kernel principal components analysis. In this brief, we propose an extension of the KECA method, named optimized KECA (OKECA), that directly extracts the optimal features retaining most of the data entropy by means of compacting the information in very few features (often in just one or two). The proposed method produces features which have higher expressive power. In particular, it is based on the independent component analysis framework, and introduces an extra rotation to the eigen decomposition, which is optimized via gradient-ascent search. This maximum entropy preservation suggests that OKECA features are more efficient than KECA features for density estimation. In addition, a critical issue in both the methods is the selection of the kernel parameter, since it critically affects the resulting performance. Here, we analyze the most common kernel length-scale selection criteria. The results of both the methods are illustrated in different synthetic and real problems. Results show that OKECA returns projections with more expressive power than KECA, the most successful rule for estimating the kernel parameter is based on maximum likelihood, and OKECA is more robust to the selection of the length-scale parameter in kernel density estimation.
Seismic noise attenuation using an online subspace tracking algorithm
Zhou, Yatong; Li, Shuhua; Zhang, Dong; Chen, Yangkang
2018-02-01
We propose a new low-rank based noise attenuation method using an efficient algorithm for tracking subspaces from highly corrupted seismic observations. The subspace tracking algorithm requires only basic linear algebraic manipulations. The algorithm is derived by analysing incremental gradient descent on the Grassmannian manifold of subspaces. When the multidimensional seismic data are mapped to a low-rank space, the subspace tracking algorithm can be directly applied to the input low-rank matrix to estimate the useful signals. Since the subspace tracking algorithm is an online algorithm, it is more robust to random noise than traditional truncated singular value decomposition (TSVD) based subspace tracking algorithm. Compared with the state-of-the-art algorithms, the proposed denoising method can obtain better performance. More specifically, the proposed method outperforms the TSVD-based singular spectrum analysis method in causing less residual noise and also in saving half of the computational cost. Several synthetic and field data examples with different levels of complexities demonstrate the effectiveness and robustness of the presented algorithm in rejecting different types of noise including random noise, spiky noise, blending noise, and coherent noise.
Mapping Fire Severity Using Imaging Spectroscopy and Kernel Based Image Analysis
Prasad, S.; Cui, M.; Zhang, Y.; Veraverbeke, S.
2014-12-01
Improved spatial representation of within-burn heterogeneity after wildfires is paramount to effective land management decisions and more accurate fire emissions estimates. In this work, we demonstrate feasibility and efficacy of airborne imaging spectroscopy (hyperspectral imagery) for quantifying wildfire burn severity, using kernel based image analysis techniques. Two different airborne hyperspectral datasets, acquired over the 2011 Canyon and 2013 Rim fire in California using the Airborne Visible InfraRed Imaging Spectrometer (AVIRIS) sensor, were used in this study. The Rim Fire, covering parts of the Yosemite National Park started on August 17, 2013, and was the third largest fire in California's history. Canyon Fire occurred in the Tehachapi mountains, and started on September 4, 2011. In addition to post-fire data for both fires, half of the Rim fire was also covered with pre-fire images. Fire severity was measured in the field using Geo Composite Burn Index (GeoCBI). The field data was utilized to train and validate our models, wherein the trained models, in conjunction with imaging spectroscopy data were used for GeoCBI estimation wide geographical regions. This work presents an approach for using remotely sensed imagery combined with GeoCBI field data to map fire scars based on a non-linear (kernel based) epsilon-Support Vector Regression (e-SVR), which was used to learn the relationship between spectra and GeoCBI in a kernel-induced feature space. Classification of healthy vegetation versus fire-affected areas based on morphological multi-attribute profiles was also studied. The availability of pre- and post-fire imaging spectroscopy data over the Rim Fire provided a unique opportunity to evaluate the performance of bi-temporal imaging spectroscopy for assessing post-fire effects. This type of data is currently constrained because of limited airborne acquisitions before a fire, but will become widespread with future spaceborne sensors such as those on
Bérenger Bramas
2018-04-01
Full Text Available The sparse matrix-vector product (SpMV is a fundamental operation in many scientific applications from various fields. The High Performance Computing (HPC community has therefore continuously invested a lot of effort to provide an efficient SpMV kernel on modern CPU architectures. Although it has been shown that block-based kernels help to achieve high performance, they are difficult to use in practice because of the zero padding they require. In the current paper, we propose new kernels using the AVX-512 instruction set, which makes it possible to use a blocking scheme without any zero padding in the matrix memory storage. We describe mask-based sparse matrix formats and their corresponding SpMV kernels highly optimized in assembly language. Considering that the optimal blocking size depends on the matrix, we also provide a method to predict the best kernel to be used utilizing a simple interpolation of results from previous executions. We compare the performance of our approach to that of the Intel MKL CSR kernel and the CSR5 open-source package on a set of standard benchmark matrices. We show that we can achieve significant improvements in many cases, both for sequential and for parallel executions. Finally, we provide the corresponding code in an open source library, called SPC5.
Efficient 3D movement-based kernel density estimator and application to wildlife ecology
Tracey-PR, Jeff; Sheppard, James K.; Lockwood, Glenn K.; Chourasia, Amit; Tatineni, Mahidhar; Fisher, Robert N.; Sinkovits, Robert S.
2014-01-01
We describe an efficient implementation of a 3D movement-based kernel density estimator for determining animal space use from discrete GPS measurements. This new method provides more accurate results, particularly for species that make large excursions in the vertical dimension. The downside of this approach is that it is much more computationally expensive than simpler, lower-dimensional models. Through a combination of code restructuring, parallelization and performance optimization, we were able to reduce the time to solution by up to a factor of 1000x, thereby greatly improving the applicability of the method.
Yi-Hung Liu
2014-07-01
Full Text Available Electroencephalogram-based emotion recognition (EEG-ER has received increasing attention in the fields of health care, affective computing, and brain-computer interface (BCI. However, satisfactory ER performance within a bi-dimensional and non-discrete emotional space using single-trial EEG data remains a challenging task. To address this issue, we propose a three-layer scheme for single-trial EEG-ER. In the first layer, a set of spectral powers of different EEG frequency bands are extracted from multi-channel single-trial EEG signals. In the second layer, the kernel Fisher’s discriminant analysis method is applied to further extract features with better discrimination ability from the EEG spectral powers. The feature vector produced by layer 2 is called a kernel Fisher’s emotion pattern (KFEP, and is sent into layer 3 for further classification where the proposed imbalanced quasiconformal kernel support vector machine (IQK-SVM serves as the emotion classifier. The outputs of the three layer EEG-ER system include labels of emotional valence and arousal. Furthermore, to collect effective training and testing datasets for the current EEG-ER system, we also use an emotion-induction paradigm in which a set of pictures selected from the International Affective Picture System (IAPS are employed as emotion induction stimuli. The performance of the proposed three-layer solution is compared with that of other EEG spectral power-based features and emotion classifiers. Results on 10 healthy participants indicate that the proposed KFEP feature performs better than other spectral power features, and IQK-SVM outperforms traditional SVM in terms of the EEG-ER accuracy. Our findings also show that the proposed EEG-ER scheme achieves the highest classification accuracies of valence (82.68% and arousal (84.79% among all testing methods.
Kernel based methods for accelerated failure time model with ultra-high dimensional data
Jiang Feng
2010-12-01
Full Text Available Abstract Background Most genomic data have ultra-high dimensions with more than 10,000 genes (probes. Regularization methods with L1 and Lp penalty have been extensively studied in survival analysis with high-dimensional genomic data. However, when the sample size n ≪ m (the number of genes, directly identifying a small subset of genes from ultra-high (m > 10, 000 dimensional data is time-consuming and not computationally efficient. In current microarray analysis, what people really do is select a couple of thousands (or hundreds of genes using univariate analysis or statistical tests, and then apply the LASSO-type penalty to further reduce the number of disease associated genes. This two-step procedure may introduce bias and inaccuracy and lead us to miss biologically important genes. Results The accelerated failure time (AFT model is a linear regression model and a useful alternative to the Cox model for survival analysis. In this paper, we propose a nonlinear kernel based AFT model and an efficient variable selection method with adaptive kernel ridge regression. Our proposed variable selection method is based on the kernel matrix and dual problem with a much smaller n × n matrix. It is very efficient when the number of unknown variables (genes is much larger than the number of samples. Moreover, the primal variables are explicitly updated and the sparsity in the solution is exploited. Conclusions Our proposed methods can simultaneously identify survival associated prognostic factors and predict survival outcomes with ultra-high dimensional genomic data. We have demonstrated the performance of our methods with both simulation and real data. The proposed method performs superbly with limited computational studies.
Xin Chen
2015-01-01
Full Text Available Adaptive Dynamic Programming (ADP with critic-actor architecture is an effective way to perform online learning control. To avoid the subjectivity in the design of a neural network that serves as a critic network, kernel-based adaptive critic design (ACD was developed recently. There are two essential issues for a static kernel-based model: how to determine proper hyperparameters in advance and how to select right samples to describe the value function. They all rely on the assessment of sample values. Based on the theoretical analysis, this paper presents a two-phase simultaneous learning method for a Gaussian-kernel-based critic network. It is able to estimate the values of samples without infinitively revisiting them. And the hyperparameters of the kernel model are optimized simultaneously. Based on the estimated sample values, the sample set can be refined by adding alternatives or deleting redundances. Combining this critic design with actor network, we present a Gaussian-kernel-based Adaptive Dynamic Programming (GK-ADP approach. Simulations are used to verify its feasibility, particularly the necessity of two-phase learning, the convergence characteristics, and the improvement of the system performance by using a varying sample set.
Wang, Gang; Wang, Yalin
2017-02-15
In this paper, we propose a heat kernel based regional shape descriptor that may be capable of better exploiting volumetric morphological information than other available methods, thereby improving statistical power on brain magnetic resonance imaging (MRI) analysis. The mechanism of our analysis is driven by the graph spectrum and the heat kernel theory, to capture the volumetric geometry information in the constructed tetrahedral meshes. In order to capture profound brain grey matter shape changes, we first use the volumetric Laplace-Beltrami operator to determine the point pair correspondence between white-grey matter and CSF-grey matter boundary surfaces by computing the streamlines in a tetrahedral mesh. Secondly, we propose multi-scale grey matter morphology signatures to describe the transition probability by random walk between the point pairs, which reflects the inherent geometric characteristics. Thirdly, a point distribution model is applied to reduce the dimensionality of the grey matter morphology signatures and generate the internal structure features. With the sparse linear discriminant analysis, we select a concise morphology feature set with improved classification accuracies. In our experiments, the proposed work outperformed the cortical thickness features computed by FreeSurfer software in the classification of Alzheimer's disease and its prodromal stage, i.e., mild cognitive impairment, on publicly available data from the Alzheimer's Disease Neuroimaging Initiative. The multi-scale and physics based volumetric structure feature may bring stronger statistical power than some traditional methods for MRI-based grey matter morphology analysis. Copyright © 2016 Elsevier Inc. All rights reserved.
Modelling lateral beam quality variations in pencil kernel based photon dose calculations
Nyholm, T; Olofsson, J; Ahnesjoe, A; Karlsson, M
2006-01-01
Standard treatment machines for external radiotherapy are designed to yield flat dose distributions at a representative treatment depth. The common method to reach this goal is to use a flattening filter to decrease the fluence in the centre of the beam. A side effect of this filtering is that the average energy of the beam is generally lower at a distance from the central axis, a phenomenon commonly referred to as off-axis softening. The off-axis softening results in a relative change in beam quality that is almost independent of machine brand and model. Central axis dose calculations using pencil beam kernels show no drastic loss in accuracy when the off-axis beam quality variations are neglected. However, for dose calculated at off-axis positions the effect should be considered, otherwise errors of several per cent can be introduced. This work proposes a method to explicitly include the effect of off-axis softening in pencil kernel based photon dose calculations for arbitrary positions in a radiation field. Variations of pencil kernel values are modelled through a generic relation between half value layer (HVL) thickness and off-axis position for standard treatment machines. The pencil kernel integration for dose calculation is performed through sampling of energy fluence and beam quality in sectors of concentric circles around the calculation point. The method is fully based on generic data and therefore does not require any specific measurements for characterization of the off-axis softening effect, provided that the machine performance is in agreement with the assumed HVL variations. The model is verified versus profile measurements at different depths and through a model self-consistency check, using the dose calculation model to estimate HVL values at off-axis positions. A comparison between calculated and measured profiles at different depths showed a maximum relative error of 4% without explicit modelling of off-axis softening. The maximum relative error
Fault Detection and Diagnosis for Gas Turbines Based on a Kernelized Information Entropy Model
Weiying Wang
2014-01-01
Full Text Available Gas turbines are considered as one kind of the most important devices in power engineering and have been widely used in power generation, airplanes, and naval ships and also in oil drilling platforms. However, they are monitored without man on duty in the most cases. It is highly desirable to develop techniques and systems to remotely monitor their conditions and analyze their faults. In this work, we introduce a remote system for online condition monitoring and fault diagnosis of gas turbine on offshore oil well drilling platforms based on a kernelized information entropy model. Shannon information entropy is generalized for measuring the uniformity of exhaust temperatures, which reflect the overall states of the gas paths of gas turbine. In addition, we also extend the entropy to compute the information quantity of features in kernel spaces, which help to select the informative features for a certain recognition task. Finally, we introduce the information entropy based decision tree algorithm to extract rules from fault samples. The experiments on some real-world data show the effectiveness of the proposed algorithms.
Fault detection and diagnosis for gas turbines based on a kernelized information entropy model.
Wang, Weiying; Xu, Zhiqiang; Tang, Rui; Li, Shuying; Wu, Wei
2014-01-01
Gas turbines are considered as one kind of the most important devices in power engineering and have been widely used in power generation, airplanes, and naval ships and also in oil drilling platforms. However, they are monitored without man on duty in the most cases. It is highly desirable to develop techniques and systems to remotely monitor their conditions and analyze their faults. In this work, we introduce a remote system for online condition monitoring and fault diagnosis of gas turbine on offshore oil well drilling platforms based on a kernelized information entropy model. Shannon information entropy is generalized for measuring the uniformity of exhaust temperatures, which reflect the overall states of the gas paths of gas turbine. In addition, we also extend the entropy to compute the information quantity of features in kernel spaces, which help to select the informative features for a certain recognition task. Finally, we introduce the information entropy based decision tree algorithm to extract rules from fault samples. The experiments on some real-world data show the effectiveness of the proposed algorithms.
Multi-class Mode of Action Classification of Toxic Compounds Using Logic Based Kernel Methods.
Lodhi, Huma; Muggleton, Stephen; Sternberg, Mike J E
2010-09-17
Toxicity prediction is essential for drug design and development of effective therapeutics. In this paper we present an in silico strategy, to identify the mode of action of toxic compounds, that is based on the use of a novel logic based kernel method. The technique uses support vector machines in conjunction with the kernels constructed from first order rules induced by an Inductive Logic Programming system. It constructs multi-class models by using a divide and conquer reduction strategy that splits multi-classes into binary groups and solves each individual problem recursively hence generating an underlying decision list structure. In order to evaluate the effectiveness of the approach for chemoinformatics problems like predictive toxicology, we apply it to toxicity classification in aquatic systems. The method is used to identify and classify 442 compounds with respect to the mode of action. The experimental results show that the technique successfully classifies toxic compounds and can be useful in assessing environmental risks. Experimental comparison of the performance of the proposed multi-class scheme with the standard multi-class Inductive Logic Programming algorithm and multi-class Support Vector Machine yields statistically significant results and demonstrates the potential power and benefits of the approach in identifying compounds of various toxic mechanisms. Copyright © 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
Xue-cun Yang
2015-01-01
Full Text Available For coal slurry pipeline blockage prediction problem, through the analysis of actual scene, it is determined that the pressure prediction from each measuring point is the premise of pipeline blockage prediction. Kernel function of support vector machine is introduced into extreme learning machine, the parameters are optimized by particle swarm algorithm, and blockage prediction method based on particle swarm optimization kernel function extreme learning machine (PSOKELM is put forward. The actual test data from HuangLing coal gangue power plant are used for simulation experiments and compared with support vector machine prediction model optimized by particle swarm algorithm (PSOSVM and kernel function extreme learning machine prediction model (KELM. The results prove that mean square error (MSE for the prediction model based on PSOKELM is 0.0038 and the correlation coefficient is 0.9955, which is superior to prediction model based on PSOSVM in speed and accuracy and superior to KELM prediction model in accuracy.
Timmerman, Marieke E.; Ceulemans, Eva; De Roover, Kim; Van Leeuwen, Karla
2013-01-01
To achieve an insightful clustering of multivariate data, we propose subspace K-means. Its central idea is to model the centroids and cluster residuals in reduced spaces, which allows for dealing with a wide range of cluster types and yields rich interpretations of the clusters. We review the
Yan, Kang K; Zhao, Hongyu; Pang, Herbert
2017-12-06
High-throughput sequencing data are widely collected and analyzed in the study of complex diseases in quest of improving human health. Well-studied algorithms mostly deal with single data source, and cannot fully utilize the potential of these multi-omics data sources. In order to provide a holistic understanding of human health and diseases, it is necessary to integrate multiple data sources. Several algorithms have been proposed so far, however, a comprehensive comparison of data integration algorithms for classification of binary traits is currently lacking. In this paper, we focus on two common classes of integration algorithms, graph-based that depict relationships with subjects denoted by nodes and relationships denoted by edges, and kernel-based that can generate a classifier in feature space. Our paper provides a comprehensive comparison of their performance in terms of various measurements of classification accuracy and computation time. Seven different integration algorithms, including graph-based semi-supervised learning, graph sharpening integration, composite association network, Bayesian network, semi-definite programming-support vector machine (SDP-SVM), relevance vector machine (RVM) and Ada-boost relevance vector machine are compared and evaluated with hypertension and two cancer data sets in our study. In general, kernel-based algorithms create more complex models and require longer computation time, but they tend to perform better than graph-based algorithms. The performance of graph-based algorithms has the advantage of being faster computationally. The empirical results demonstrate that composite association network, relevance vector machine, and Ada-boost RVM are the better performers. We provide recommendations on how to choose an appropriate algorithm for integrating data from multiple sources.
Yekkehkhany, B.; Safari, A.; Homayouni, S.; Hasanlou, M.
2014-10-01
In this paper, a framework is developed based on Support Vector Machines (SVM) for crop classification using polarimetric features extracted from multi-temporal Synthetic Aperture Radar (SAR) imageries. The multi-temporal integration of data not only improves the overall retrieval accuracy but also provides more reliable estimates with respect to single-date data. Several kernel functions are employed and compared in this study for mapping the input space to higher Hilbert dimension space. These kernel functions include linear, polynomials and Radial Based Function (RBF). The method is applied to several UAVSAR L-band SAR images acquired over an agricultural area near Winnipeg, Manitoba, Canada. In this research, the temporal alpha features of H/A/α decomposition method are used in classification. The experimental tests show an SVM classifier with RBF kernel for three dates of data increases the Overall Accuracy (OA) to up to 3% in comparison to using linear kernel function, and up to 1% in comparison to a 3rd degree polynomial kernel function.
Pencil kernel correction and residual error estimation for quality-index-based dose calculations
Nyholm, Tufve; Olofsson, Joergen; Ahnesjoe, Anders; Georg, Dietmar; Karlsson, Mikael
2006-01-01
Experimental data from 593 photon beams were used to quantify the errors in dose calculations using a previously published pencil kernel model. A correction of the kernel was derived in order to remove the observed systematic errors. The remaining residual error for individual beams was modelled through uncertainty associated with the kernel model. The methods were tested against an independent set of measurements. No significant systematic error was observed in the calculations using the derived correction of the kernel and the remaining random errors were found to be adequately predicted by the proposed method
Two-qubit quantum computing in a projected subspace
Bi Qiao; Ruda, H.E.; Zhan, M.S.
2002-01-01
A formulation for performing quantum computing in a projected subspace is presented, based on the subdynamical kinetic equation (SKE) for an open quantum system. The eigenvectors of the kinetic equation are shown to remain invariant before and after interaction with the environment. However, the eigenvalues in the projected subspace exhibit a type of phase shift to the evolutionary states. This phase shift does not destroy the decoherence-free (DF) property of the subspace because the associated fidelity is 1. This permits a universal formalism to be presented--the eigenprojectors of the free part of the Hamiltonian for the system and bath may be used to construct a DF projected subspace based on the SKE. To eliminate possible phase or unitary errors induced by the change in the eigenvalues, a cancellation technique is proposed, using the adjustment of the coupling time, and applied to a two-qubit computing system. A general criteria for constructing a DF-projected subspace from the SKE is discussed. Finally, a proposal for using triangulation to realize a decoherence-free subsystem based on SKE is presented. The concrete formulation for a two-qubit model is given exactly. Our approach is general and appears to be applicable to any type of decoherence
Palmprint and Face Multi-Modal Biometric Recognition Based on SDA-GSVD and Its Kernelization
Jing-Yu Yang
2012-04-01
Full Text Available When extracting discriminative features from multimodal data, current methods rarely concern themselves with the data distribution. In this paper, we present an assumption that is consistent with the viewpoint of discrimination, that is, a person’s overall biometric data should be regarded as one class in the input space, and his different biometric data can form different Gaussians distributions, i.e., different subclasses. Hence, we propose a novel multimodal feature extraction and recognition approach based on subclass discriminant analysis (SDA. Specifically, one person’s different bio-data are treated as different subclasses of one class, and a transformed space is calculated, where the difference among subclasses belonging to different persons is maximized, and the difference within each subclass is minimized. Then, the obtained multimodal features are used for classification. Two solutions are presented to overcome the singularity problem encountered in calculation, which are using PCA preprocessing, and employing the generalized singular value decomposition (GSVD technique, respectively. Further, we provide nonlinear extensions of SDA based multimodal feature extraction, that is, the feature fusion based on KPCA-SDA and KSDA-GSVD. In KPCA-SDA, we first apply Kernel PCA on each single modal before performing SDA. While in KSDA-GSVD, we directly perform Kernel SDA to fuse multimodal data by applying GSVD to avoid the singular problem. For simplicity two typical types of biometric data are considered in this paper, i.e., palmprint data and face data. Compared with several representative multimodal biometrics recognition methods, experimental results show that our approaches outperform related multimodal recognition methods and KSDA-GSVD achieves the best recognition performance.
Palmprint and face multi-modal biometric recognition based on SDA-GSVD and its kernelization.
Jing, Xiao-Yuan; Li, Sheng; Li, Wen-Qian; Yao, Yong-Fang; Lan, Chao; Lu, Jia-Sen; Yang, Jing-Yu
2012-01-01
When extracting discriminative features from multimodal data, current methods rarely concern themselves with the data distribution. In this paper, we present an assumption that is consistent with the viewpoint of discrimination, that is, a person's overall biometric data should be regarded as one class in the input space, and his different biometric data can form different Gaussians distributions, i.e., different subclasses. Hence, we propose a novel multimodal feature extraction and recognition approach based on subclass discriminant analysis (SDA). Specifically, one person's different bio-data are treated as different subclasses of one class, and a transformed space is calculated, where the difference among subclasses belonging to different persons is maximized, and the difference within each subclass is minimized. Then, the obtained multimodal features are used for classification. Two solutions are presented to overcome the singularity problem encountered in calculation, which are using PCA preprocessing, and employing the generalized singular value decomposition (GSVD) technique, respectively. Further, we provide nonlinear extensions of SDA based multimodal feature extraction, that is, the feature fusion based on KPCA-SDA and KSDA-GSVD. In KPCA-SDA, we first apply Kernel PCA on each single modal before performing SDA. While in KSDA-GSVD, we directly perform Kernel SDA to fuse multimodal data by applying GSVD to avoid the singular problem. For simplicity two typical types of biometric data are considered in this paper, i.e., palmprint data and face data. Compared with several representative multimodal biometrics recognition methods, experimental results show that our approaches outperform related multimodal recognition methods and KSDA-GSVD achieves the best recognition performance.
Palmprint and Face Multi-Modal Biometric Recognition Based on SDA-GSVD and Its Kernelization
Jing, Xiao-Yuan; Li, Sheng; Li, Wen-Qian; Yao, Yong-Fang; Lan, Chao; Lu, Jia-Sen; Yang, Jing-Yu
2012-01-01
When extracting discriminative features from multimodal data, current methods rarely concern themselves with the data distribution. In this paper, we present an assumption that is consistent with the viewpoint of discrimination, that is, a person's overall biometric data should be regarded as one class in the input space, and his different biometric data can form different Gaussians distributions, i.e., different subclasses. Hence, we propose a novel multimodal feature extraction and recognition approach based on subclass discriminant analysis (SDA). Specifically, one person's different bio-data are treated as different subclasses of one class, and a transformed space is calculated, where the difference among subclasses belonging to different persons is maximized, and the difference within each subclass is minimized. Then, the obtained multimodal features are used for classification. Two solutions are presented to overcome the singularity problem encountered in calculation, which are using PCA preprocessing, and employing the generalized singular value decomposition (GSVD) technique, respectively. Further, we provide nonlinear extensions of SDA based multimodal feature extraction, that is, the feature fusion based on KPCA-SDA and KSDA-GSVD. In KPCA-SDA, we first apply Kernel PCA on each single modal before performing SDA. While in KSDA-GSVD, we directly perform Kernel SDA to fuse multimodal data by applying GSVD to avoid the singular problem. For simplicity two typical types of biometric data are considered in this paper, i.e., palmprint data and face data. Compared with several representative multimodal biometrics recognition methods, experimental results show that our approaches outperform related multimodal recognition methods and KSDA-GSVD achieves the best recognition performance. PMID:22778600
Subspace preservation, subspace locality, and gluing of completely positive maps
Aaberg, Johan
2004-01-01
Three concepts concerning completely positive maps (CPMs) and trace preserving CPMs (channels) are introduced and investigated. These are named subspace preserving (SP) CPMs, subspace local (SL) channels, and gluing of CPMs. SP CPMs has, in the case of trace preserving CPMs, a simple interpretation as those which preserve probability weights on a given orthogonal sum decomposition of the Hilbert space of a quantum system. The proposed definition of subspace locality of quantum channels is an attempt to answer the question of what kind of restriction should be put on a channel, if it is to act 'locally' with respect to two 'locations', when these naturally correspond to a separation of the total Hilbert space in an orthogonal sum of subspaces, rather than a tensor product decomposition. As a description of the concept of gluings of quantum channels, consider a pair of 'evolution machines', each with the ability to evolve the internal state of a 'particle' inserted into its input. Each of these machines is characterized by a channel describing the operation the internal state has experienced when the particle is returned at the output. Suppose a particle is put in a superposition between the input of the first and the second machine. Here it is shown that the total evolution caused by a pair of such devices is not uniquely determined by the channels of the two machines. Such 'global' channels describing the machine pair are examples of gluings of the two single machine channels. Various expressions to generate the set of SP and SL channels, as well as expressions to generate the set of gluings of given channels, are deduced. We discuss conceptual aspects of the nature of these channels and the nature of the non-uniqueness of gluings
Generic primal-dual interior point methods based on a new kernel function
EL Ghami, M.; Roos, C.
2008-01-01
In this paper we present a generic primal-dual interior point methods (IPMs) for linear optimization in which the search direction depends on a univariate kernel function which is also used as proximity measure in the analysis of the algorithm. The proposed kernel function does not satisfy all the
Ensemble-based forecasting at Horns Rev: Ensemble conversion and kernel dressing
Pinson, Pierre; Madsen, Henrik
. The obtained ensemble forecasts of wind power are then converted into predictive distributions with an original adaptive kernel dressing method. The shape of the kernels is driven by a mean-variance model, the parameters of which are recursively estimated in order to maximize the overall skill of obtained...
Improving Change Detection in Forest Areas Based on Stereo Panchromatic Imagery Using Kernel MNF
Tian, Jiaojiao; Nielsen, Allan Aasbjerg; Reinartz, Peter
2014-01-01
with other unrelated phenomena, e.g., seasonal changes of land covers such as grass and crops. Therefore, we propose an approach that exploits kernel Minimum Noise Fraction (kMNF) to transform simple change features into high-dimensional feature space. Digital surface models (DSMs) generated from stereo...... imagery are used to provide information on height difference, which is additionally used to separate forest changes from other land-cover changes. With very few training samples, a change mask is generated with iterated canonical discriminant analysis (ICDA). Two examples are presented to illustrate...... the approach and demonstrate its efficiency. It is shown that with the same amount of training samples, the proposed method can obtain more accurate change masks compared with algorithms based on k-means, one-class support vector machine, and random forests....
Sparse approximation of multilinear problems with applications to kernel-based methods in UQ
Nobile, Fabio; Tempone, Raul; Wolfers, Sö ren
2017-01-01
We provide a framework for the sparse approximation of multilinear problems and show that several problems in uncertainty quantification fit within this framework. In these problems, the value of a multilinear map has to be approximated using approximations of different accuracy and computational work of the arguments of this map. We propose and analyze a generalized version of Smolyak’s algorithm, which provides sparse approximation formulas with convergence rates that mitigate the curse of dimension that appears in multilinear approximation problems with a large number of arguments. We apply the general framework to response surface approximation and optimization under uncertainty for parametric partial differential equations using kernel-based approximation. The theoretical results are supplemented by numerical experiments.
Sparse approximation of multilinear problems with applications to kernel-based methods in UQ
Nobile, Fabio
2017-11-16
We provide a framework for the sparse approximation of multilinear problems and show that several problems in uncertainty quantification fit within this framework. In these problems, the value of a multilinear map has to be approximated using approximations of different accuracy and computational work of the arguments of this map. We propose and analyze a generalized version of Smolyak’s algorithm, which provides sparse approximation formulas with convergence rates that mitigate the curse of dimension that appears in multilinear approximation problems with a large number of arguments. We apply the general framework to response surface approximation and optimization under uncertainty for parametric partial differential equations using kernel-based approximation. The theoretical results are supplemented by numerical experiments.
Fernández-Delgado, Manuel; Cernadas, Eva; Barro, Senén; Ribeiro, Jorge; Neves, José
2014-02-01
The Direct Kernel Perceptron (DKP) (Fernández-Delgado et al., 2010) is a very simple and fast kernel-based classifier, related to the Support Vector Machine (SVM) and to the Extreme Learning Machine (ELM) (Huang, Wang, & Lan, 2011), whose α-coefficients are calculated directly, without any iterative training, using an analytical closed-form expression which involves only the training patterns. The DKP, which is inspired by the Direct Parallel Perceptron, (Auer et al., 2008), uses a Gaussian kernel and a linear classifier (perceptron). The weight vector of this classifier in the feature space minimizes an error measure which combines the training error and the hyperplane margin, without any tunable regularization parameter. This weight vector can be translated, using a variable change, to the α-coefficients, and both are determined without iterative calculations. We calculate solutions using several error functions, achieving the best trade-off between accuracy and efficiency with the linear function. These solutions for the α coefficients can be considered alternatives to the ELM with a new physical meaning in terms of error and margin: in fact, the linear and quadratic DKP are special cases of the two-class ELM when the regularization parameter C takes the values C=0 and C=∞. The linear DKP is extremely efficient and much faster (over a vast collection of 42 benchmark and real-life data sets) than 12 very popular and accurate classifiers including SVM, Multi-Layer Perceptron, Adaboost, Random Forest and Bagging of RPART decision trees, Linear Discriminant Analysis, K-Nearest Neighbors, ELM, Probabilistic Neural Networks, Radial Basis Function neural networks and Generalized ART. Besides, despite its simplicity and extreme efficiency, DKP achieves higher accuracies than 7 out of 12 classifiers, exhibiting small differences with respect to the best ones (SVM, ELM, Adaboost and Random Forest), which are much slower. Thus, the DKP provides an easy and fast way
A kernel-based multivariate feature selection method for microarray data classification.
Shiquan Sun
Full Text Available High dimensionality and small sample sizes, and their inherent risk of overfitting, pose great challenges for constructing efficient classifiers in microarray data classification. Therefore a feature selection technique should be conducted prior to data classification to enhance prediction performance. In general, filter methods can be considered as principal or auxiliary selection mechanism because of their simplicity, scalability, and low computational complexity. However, a series of trivial examples show that filter methods result in less accurate performance because they ignore the dependencies of features. Although few publications have devoted their attention to reveal the relationship of features by multivariate-based methods, these methods describe relationships among features only by linear methods. While simple linear combination relationship restrict the improvement in performance. In this paper, we used kernel method to discover inherent nonlinear correlations among features as well as between feature and target. Moreover, the number of orthogonal components was determined by kernel Fishers linear discriminant analysis (FLDA in a self-adaptive manner rather than by manual parameter settings. In order to reveal the effectiveness of our method we performed several experiments and compared the results between our method and other competitive multivariate-based features selectors. In our comparison, we used two classifiers (support vector machine, [Formula: see text]-nearest neighbor on two group datasets, namely two-class and multi-class datasets. Experimental results demonstrate that the performance of our method is better than others, especially on three hard-classify datasets, namely Wang's Breast Cancer, Gordon's Lung Adenocarcinoma and Pomeroy's Medulloblastoma.
Adiabatic evolution of decoherence-free subspaces and its shortcuts
Wu, S. L.; Huang, X. L.; Li, H.; Yi, X. X.
2017-10-01
The adiabatic theorem and shortcuts to adiabaticity for time-dependent open quantum systems are explored in this paper. Starting from the definition of dynamical stable decoherence-free subspace, we show that, under a compact adiabatic condition, the quantum state remains in the time-dependent decoherence-free subspace with an extremely high purity, even though the dynamics of the open quantum system may not be adiabatic. The adiabatic condition mentioned here in the adiabatic theorem for open systems is very similar to that for closed quantum systems, except that the operators required to change slowly are the Lindblad operators. We also show that the adiabatic evolution of decoherence-free subspaces depends on the existence of instantaneous decoherence-free subspaces, which requires that the Hamiltonian of open quantum systems be engineered according to the incoherent control protocol. In addition, shortcuts to adiabaticity for adiabatic decoherence-free subspaces are also presented based on the transitionless quantum driving method. Finally, we provide an example that consists of a two-level system coupled to a broadband squeezed vacuum field to show our theory. Our approach employs Markovian master equations and the theory can apply to finite-dimensional quantum open systems.
Rizvi, S.Z.; Mohammadpour, J.; Toth, R.; Meskin, N.
2015-01-01
This paper first describes the development of a nonparametric identification method for linear parameter-varying (LPV) state-space models and then applies it to a nonlinear process system. The proposed method uses kernel-based least-squares support vector machines (LS-SVM). While parametric
Lippert, Christoph; Xiang, Jing; Horta, Danilo; Widmer, Christian; Kadie, Carl; Heckerman, David; Listgarten, Jennifer
2014-11-15
Set-based variance component tests have been identified as a way to increase power in association studies by aggregating weak individual effects. However, the choice of test statistic has been largely ignored even though it may play an important role in obtaining optimal power. We compared a standard statistical test-a score test-with a recently developed likelihood ratio (LR) test. Further, when correction for hidden structure is needed, or gene-gene interactions are sought, state-of-the art algorithms for both the score and LR tests can be computationally impractical. Thus we develop new computationally efficient methods. After reviewing theoretical differences in performance between the score and LR tests, we find empirically on real data that the LR test generally has more power. In particular, on 15 of 17 real datasets, the LR test yielded at least as many associations as the score test-up to 23 more associations-whereas the score test yielded at most one more association than the LR test in the two remaining datasets. On synthetic data, we find that the LR test yielded up to 12% more associations, consistent with our results on real data, but also observe a regime of extremely small signal where the score test yielded up to 25% more associations than the LR test, consistent with theory. Finally, our computational speedups now enable (i) efficient LR testing when the background kernel is full rank, and (ii) efficient score testing when the background kernel changes with each test, as for gene-gene interaction tests. The latter yielded a factor of 2000 speedup on a cohort of size 13 500. Software available at http://research.microsoft.com/en-us/um/redmond/projects/MSCompBio/Fastlmm/. heckerma@microsoft.com Supplementary data are available at Bioinformatics online. © The Author 2014. Published by Oxford University Press.
Subspace Arrangement Codes and Cryptosystems
2011-05-09
Signature Date Acceptance for the Trident Scholar Committee Professor Carl E. Wick Associate Director of Midshipmen Research Signature Date SUBSPACE...Professor William Traves. I also thank Professor Carl Wick and the Trident Scholar Committee for providing me with the opportunity to conduct this... Sagan . Why the characteristic polynomial factors. Bulletin of the American Mathematical Society, 36(2):113–133, February 1999. [16] Karen E. Smith
Bochev, Mikhail A.; Oseledets, I.V.; Tyrtyshnikov, E.E.
2013-01-01
The aim of this paper is two-fold. First, we propose an efficient implementation of the continuous time waveform relaxation method based on block Krylov subspaces. Second, we compare this new implementation against Krylov subspace methods combined with the shift and invert technique.
Subspace Correction Methods for Total Variation and $\\ell_1$-Minimization
Fornasier, Massimo
2009-01-01
This paper is concerned with the numerical minimization of energy functionals in Hilbert spaces involving convex constraints coinciding with a seminorm for a subspace. The optimization is realized by alternating minimizations of the functional on a sequence of orthogonal subspaces. On each subspace an iterative proximity-map algorithm is implemented via oblique thresholding, which is the main new tool introduced in this work. We provide convergence conditions for the algorithm in order to compute minimizers of the target energy. Analogous results are derived for a parallel variant of the algorithm. Applications are presented in domain decomposition methods for degenerate elliptic PDEs arising in total variation minimization and in accelerated sparse recovery algorithms based on 1-minimization. We include numerical examples which show e.cient solutions to classical problems in signal and image processing. © 2009 Society for Industrial and Applied Physics.
Krylov subspace methods for solving large unsymmetric linear systems
Saad, Y.
1981-01-01
Some algorithms based upon a projection process onto the Krylov subspace K/sub m/ = Span(r 0 , Ar 0 ,...,A/sup m/-1r 0 ) are developed, generalizing the method of conjugate gradients to unsymmetric systems. These methods are extensions of Arnoldi's algorithm for solving eigenvalue problems. The convergence is analyzed in terms of the distance of the solution to the subspace K/sub m/ and some error bounds are established showing, in particular, a similarity with the conjugate gradient method (for symmetric matrices) when the eigenvalues are real. Several numerical experiments are described and discussed
A subspace preconditioning algorithm for eigenvector/eigenvalue computation
Bramble, J.H.; Knyazev, A.V.; Pasciak, J.E.
1996-12-31
We consider the problem of computing a modest number of the smallest eigenvalues along with orthogonal bases for the corresponding eigen-spaces of a symmetric positive definite matrix. In our applications, the dimension of a matrix is large and the cost of its inverting is prohibitive. In this paper, we shall develop an effective parallelizable technique for computing these eigenvalues and eigenvectors utilizing subspace iteration and preconditioning. Estimates will be provided which show that the preconditioned method converges linearly and uniformly in the matrix dimension when used with a uniform preconditioner under the assumption that the approximating subspace is close enough to the span of desired eigenvectors.
Yongliang Lin
2016-10-01
Full Text Available In this paper, we propose a multiple kernel relevance vector machine (RVM method based on the adaptive cloud particle swarm optimization (PSO algorithm to map landslide susceptibility in the low hill area of Sichuan Province, China. In the multi-kernel structure, the kernel selection problem can be solved by adjusting the kernel weight, which determines the single kernel contribution of the final kernel mapping. The weights and parameters of the multi-kernel function were optimized using the PSO algorithm. In addition, the convergence speed of the PSO algorithm was increased using cloud theory. To ensure the stability of the prediction model, the result of a five-fold cross-validation method was used as the fitness of the PSO algorithm. To verify the results, receiver operating characteristic curves (ROC and landslide dot density (LDD were used. The results show that the model that used a heterogeneous kernel (a combination of two different kernel functions had a larger area under the ROC curve (0.7616 and a lower prediction error ratio (0.28% than did the other types of kernel models employed in this study. In addition, both the sum of two high susceptibility zone LDDs (6.71/100 km2 and the sum of two low susceptibility zone LDDs (0.82/100 km2 demonstrated that the landslide susceptibility map based on the heterogeneous kernel model was closest to the historical landslide distribution. In conclusion, the results obtained in this study can provide very useful information for disaster prevention and land-use planning in the study area.
Kernel learning algorithms for face recognition
Li, Jun-Bao; Pan, Jeng-Shyang
2013-01-01
Kernel Learning Algorithms for Face Recognition covers the framework of kernel based face recognition. This book discusses the advanced kernel learning algorithms and its application on face recognition. This book also focuses on the theoretical deviation, the system framework and experiments involving kernel based face recognition. Included within are algorithms of kernel based face recognition, and also the feasibility of the kernel based face recognition method. This book provides researchers in pattern recognition and machine learning area with advanced face recognition methods and its new
Kernel density estimation-based real-time prediction for respiratory motion
Ruan, Dan
2010-01-01
Effective delivery of adaptive radiotherapy requires locating the target with high precision in real time. System latency caused by data acquisition, streaming, processing and delivery control necessitates prediction. Prediction is particularly challenging for highly mobile targets such as thoracic and abdominal tumors undergoing respiration-induced motion. The complexity of the respiratory motion makes it difficult to build and justify explicit models. In this study, we honor the intrinsic uncertainties in respiratory motion and propose a statistical treatment of the prediction problem. Instead of asking for a deterministic covariate-response map and a unique estimate value for future target position, we aim to obtain a distribution of the future target position (response variable) conditioned on the observed historical sample values (covariate variable). The key idea is to estimate the joint probability distribution (pdf) of the covariate and response variables using an efficient kernel density estimation method. Then, the problem of identifying the distribution of the future target position reduces to identifying the section in the joint pdf based on the observed covariate. Subsequently, estimators are derived based on this estimated conditional distribution. This probabilistic perspective has some distinctive advantages over existing deterministic schemes: (1) it is compatible with potentially inconsistent training samples, i.e., when close covariate variables correspond to dramatically different response values; (2) it is not restricted by any prior structural assumption on the map between the covariate and the response; (3) the two-stage setup allows much freedom in choosing statistical estimates and provides a full nonparametric description of the uncertainty for the resulting estimate. We evaluated the prediction performance on ten patient RPM traces, using the root mean squared difference between the prediction and the observed value normalized by the
Novel vehicle detection system based on stacked DoG kernel and AdaBoost
Kang, Hyun Ho; Lee, Seo Won; You, Sung Hyun
2018-01-01
This paper proposes a novel vehicle detection system that can overcome some limitations of typical vehicle detection systems using AdaBoost-based methods. The performance of the AdaBoost-based vehicle detection system is dependent on its training data. Thus, its performance decreases when the shape of a target differs from its training data, or the pattern of a preceding vehicle is not visible in the image due to the light conditions. A stacked Difference of Gaussian (DoG)–based feature extraction algorithm is proposed to address this issue by recognizing common characteristics, such as the shadow and rear wheels beneath vehicles—of vehicles under various conditions. The common characteristics of vehicles are extracted by applying the stacked DoG shaped kernel obtained from the 3D plot of an image through a convolution method and investigating only certain regions that have a similar patterns. A new vehicle detection system is constructed by combining the novel stacked DoG feature extraction algorithm with the AdaBoost method. Experiments are provided to demonstrate the effectiveness of the proposed vehicle detection system under different conditions. PMID:29513727
Azarnavid, Babak; Parand, Kourosh; Abbasbandy, Saeid
2018-06-01
This article discusses an iterative reproducing kernel method with respect to its effectiveness and capability of solving a fourth-order boundary value problem with nonlinear boundary conditions modeling beams on elastic foundations. Since there is no method of obtaining reproducing kernel which satisfies nonlinear boundary conditions, the standard reproducing kernel methods cannot be used directly to solve boundary value problems with nonlinear boundary conditions as there is no knowledge about the existence and uniqueness of the solution. The aim of this paper is, therefore, to construct an iterative method by the use of a combination of reproducing kernel Hilbert space method and a shooting-like technique to solve the mentioned problems. Error estimation for reproducing kernel Hilbert space methods for nonlinear boundary value problems have yet to be discussed in the literature. In this paper, we present error estimation for the reproducing kernel method to solve nonlinear boundary value problems probably for the first time. Some numerical results are given out to demonstrate the applicability of the method.
Controllable Subspaces of Open Quantum Dynamical Systems
Zhang Ming; Gong Erling; Xie Hongwei; Hu Dewen; Dai Hongyi
2008-01-01
This paper discusses the concept of controllable subspace for open quantum dynamical systems. It is constructively demonstrated that combining structural features of decoherence-free subspaces with the ability to perform open-loop coherent control on open quantum systems will allow decoherence-free subspaces to be controllable. This is in contrast to the observation that open quantum dynamical systems are not open-loop controllable. To a certain extent, this paper gives an alternative control theoretical interpretation on why decoherence-free subspaces can be useful for quantum computation.
Elnasir, Selma; Shamsuddin, Siti Mariyam; Farokhi, Sajad
2015-01-01
Palm vein recognition (PVR) is a promising new biometric that has been applied successfully as a method of access control by many organizations, which has even further potential in the field of forensics. The palm vein pattern has highly discriminative features that are difficult to forge because of its subcutaneous position in the palm. Despite considerable progress and a few practical issues, providing accurate palm vein readings has remained an unsolved issue in biometrics. We propose a robust and more accurate PVR method based on the combination of wavelet scattering (WS) with spectral regression kernel discriminant analysis (SRKDA). As the dimension of WS generated features is quite large, SRKDA is required to reduce the extracted features to enhance the discrimination. The results based on two public databases-PolyU Hyper Spectral Palmprint public database and PolyU Multi Spectral Palmprint-show the high performance of the proposed scheme in comparison with state-of-the-art methods. The proposed approach scored a 99.44% identification rate and a 99.90% verification rate [equal error rate (EER)=0.1%] for the hyperspectral database and a 99.97% identification rate and a 99.98% verification rate (EER=0.019%) for the multispectral database.
Kan Li
2018-04-01
Full Text Available This paper presents a novel real-time dynamic framework for quantifying time-series structure in spoken words using spikes. Audio signals are converted into multi-channel spike trains using a biologically-inspired leaky integrate-and-fire (LIF spike generator. These spike trains are mapped into a function space of infinite dimension, i.e., a Reproducing Kernel Hilbert Space (RKHS using point-process kernels, where a state-space model learns the dynamics of the multidimensional spike input using gradient descent learning. This kernelized recurrent system is very parsimonious and achieves the necessary memory depth via feedback of its internal states when trained discriminatively, utilizing the full context of the phoneme sequence. A main advantage of modeling nonlinear dynamics using state-space trajectories in the RKHS is that it imposes no restriction on the relationship between the exogenous input and its internal state. We are free to choose the input representation with an appropriate kernel, and changing the kernel does not impact the system nor the learning algorithm. Moreover, we show that this novel framework can outperform both traditional hidden Markov model (HMM speech processing as well as neuromorphic implementations based on spiking neural network (SNN, yielding accurate and ultra-low power word spotters. As a proof of concept, we demonstrate its capabilities using the benchmark TI-46 digit corpus for isolated-word automatic speech recognition (ASR or keyword spotting. Compared to HMM using Mel-frequency cepstral coefficient (MFCC front-end without time-derivatives, our MFCC-KAARMA offered improved performance. For spike-train front-end, spike-KAARMA also outperformed state-of-the-art SNN solutions. Furthermore, compared to MFCCs, spike trains provided enhanced noise robustness in certain low signal-to-noise ratio (SNR regime.
Li, Kan; Príncipe, José C
2018-01-01
This paper presents a novel real-time dynamic framework for quantifying time-series structure in spoken words using spikes. Audio signals are converted into multi-channel spike trains using a biologically-inspired leaky integrate-and-fire (LIF) spike generator. These spike trains are mapped into a function space of infinite dimension, i.e., a Reproducing Kernel Hilbert Space (RKHS) using point-process kernels, where a state-space model learns the dynamics of the multidimensional spike input using gradient descent learning. This kernelized recurrent system is very parsimonious and achieves the necessary memory depth via feedback of its internal states when trained discriminatively, utilizing the full context of the phoneme sequence. A main advantage of modeling nonlinear dynamics using state-space trajectories in the RKHS is that it imposes no restriction on the relationship between the exogenous input and its internal state. We are free to choose the input representation with an appropriate kernel, and changing the kernel does not impact the system nor the learning algorithm. Moreover, we show that this novel framework can outperform both traditional hidden Markov model (HMM) speech processing as well as neuromorphic implementations based on spiking neural network (SNN), yielding accurate and ultra-low power word spotters. As a proof of concept, we demonstrate its capabilities using the benchmark TI-46 digit corpus for isolated-word automatic speech recognition (ASR) or keyword spotting. Compared to HMM using Mel-frequency cepstral coefficient (MFCC) front-end without time-derivatives, our MFCC-KAARMA offered improved performance. For spike-train front-end, spike-KAARMA also outperformed state-of-the-art SNN solutions. Furthermore, compared to MFCCs, spike trains provided enhanced noise robustness in certain low signal-to-noise ratio (SNR) regime.
Lie n-derivations on 7 -subspace lattice algebras
all x ∈ K and all A ∈ Alg L. Based on this result, a complete characterization of linear n-Lie derivations on Alg L is obtained. Keywords. J -subspace lattice algebras; Lie derivations; Lie n-derivations; derivations. 2010 Mathematics Subject Classification. 47B47, 47L35. 1. Introduction. Let A be an algebra. Recall that a linear ...
Kernel Machine SNP-set Testing under Multiple Candidate Kernels
Wu, Michael C.; Maity, Arnab; Lee, Seunggeun; Simmons, Elizabeth M.; Harmon, Quaker E.; Lin, Xinyi; Engel, Stephanie M.; Molldrem, Jeffrey J.; Armistead, Paul M.
2013-01-01
Joint testing for the cumulative effect of multiple single nucleotide polymorphisms grouped on the basis of prior biological knowledge has become a popular and powerful strategy for the analysis of large scale genetic association studies. The kernel machine (KM) testing framework is a useful approach that has been proposed for testing associations between multiple genetic variants and many different types of complex traits by comparing pairwise similarity in phenotype between subjects to pairwise similarity in genotype, with similarity in genotype defined via a kernel function. An advantage of the KM framework is its flexibility: choosing different kernel functions allows for different assumptions concerning the underlying model and can allow for improved power. In practice, it is difficult to know which kernel to use a priori since this depends on the unknown underlying trait architecture and selecting the kernel which gives the lowest p-value can lead to inflated type I error. Therefore, we propose practical strategies for KM testing when multiple candidate kernels are present based on constructing composite kernels and based on efficient perturbation procedures. We demonstrate through simulations and real data applications that the procedures protect the type I error rate and can lead to substantially improved power over poor choices of kernels and only modest differences in power versus using the best candidate kernel. PMID:23471868
Jiang Li; Shi Tielin; Xuan Jianping
2012-01-01
Generally, the vibration signals of fault bearings are non-stationary and highly nonlinear under complicated operating conditions. Thus, it's a big challenge to extract optimal features for improving classification and simultaneously decreasing feature dimension. Kernel Marginal Fisher analysis (KMFA) is a novel supervised manifold learning algorithm for feature extraction and dimensionality reduction. In order to avoid the small sample size problem in KMFA, we propose regularized KMFA (RKMFA). A simple and efficient intelligent fault diagnosis method based on RKMFA is put forward and applied to fault recognition of rolling bearings. So as to directly excavate nonlinear features from the original high-dimensional vibration signals, RKMFA constructs two graphs describing the intra-class compactness and the inter-class separability, by combining traditional manifold learning algorithm with fisher criteria. Therefore, the optimal low-dimensional features are obtained for better classification and finally fed into the simplest K-nearest neighbor (KNN) classifier to recognize different fault categories of bearings. The experimental results demonstrate that the proposed approach improves the fault classification performance and outperforms the other conventional approaches.
A Novel Kernel-Based Regularization Technique for PET Image Reconstruction
Abdelwahhab Boudjelal
2017-06-01
Full Text Available Positron emission tomography (PET is an imaging technique that generates 3D detail of physiological processes at the cellular level. The technique requires a radioactive tracer, which decays and releases a positron that collides with an electron; consequently, annihilation photons are emitted, which can be measured. The purpose of PET is to use the measurement of photons to reconstruct the distribution of radioisotopes in the body. Currently, PET is undergoing a revamp, with advancements in data measurement instruments and the computing methods used to create the images. These computer methods are required to solve the inverse problem of “image reconstruction from projection”. This paper proposes a novel kernel-based regularization technique for maximum-likelihood expectation-maximization ( κ -MLEM to reconstruct the image. Compared to standard MLEM, the proposed algorithm is more robust and is more effective in removing background noise, whilst preserving the edges; this suppresses image artifacts, such as out-of-focus slice blur.
Model selection for Gaussian kernel PCA denoising
Jørgensen, Kasper Winther; Hansen, Lars Kai
2012-01-01
We propose kernel Parallel Analysis (kPA) for automatic kernel scale and model order selection in Gaussian kernel PCA. Parallel Analysis [1] is based on a permutation test for covariance and has previously been applied for model order selection in linear PCA, we here augment the procedure to also...... tune the Gaussian kernel scale of radial basis function based kernel PCA.We evaluate kPA for denoising of simulated data and the US Postal data set of handwritten digits. We find that kPA outperforms other heuristics to choose the model order and kernel scale in terms of signal-to-noise ratio (SNR...
A Fourier-series-based kernel-independent fast multipole method
Zhang Bo; Huang Jingfang; Pitsianis, Nikos P.; Sun Xiaobai
2011-01-01
We present in this paper a new kernel-independent fast multipole method (FMM), named as FKI-FMM, for pairwise particle interactions with translation-invariant kernel functions. FKI-FMM creates, using numerical techniques, sufficiently accurate and compressive representations of a given kernel function over multi-scale interaction regions in the form of a truncated Fourier series. It provides also economic operators for the multipole-to-multipole, multipole-to-local, and local-to-local translations that are typical and essential in the FMM algorithms. The multipole-to-local translation operator, in particular, is readily diagonal and does not dominate in arithmetic operations. FKI-FMM provides an alternative and competitive option, among other kernel-independent FMM algorithms, for an efficient application of the FMM, especially for applications where the kernel function consists of multi-physics and multi-scale components as those arising in recent studies of biological systems. We present the complexity analysis and demonstrate with experimental results the FKI-FMM performance in accuracy and efficiency.
Enzymatic Synthesis of Fatty Hydroxamic Acid Derivatives Based on Palm Kernel Oil
Sidik Silong
2011-08-01
Full Text Available Fatty hydroxamic acid derivatives were synthesized using Lipozyme TL IM catalyst at biphasic medium as the palm kernel oil was dissolved in hexane and hydroxylamine derivatives were dissolved in water: (1 N-methyl fatty hydroxamic acids (MFHAs; (2 N-isopropyl fatty hydroxamic acids (IPFHAs and (3 N-benzyl fatty hydroxamic acids (BFHAs were synthesized by reaction of palm kernel oil and N-methyl hydroxylamine (N-MHA, N-isopropyl hydroxylamine (N-IPHA and N-benzyl hydroxylamine (N-BHA, respectively. Finally, after separation the products were characterized by color testing, elemental analysis, FT-IR and 1H-NMR spectroscopy. For achieving the highest conversion percentage of product the optimum molar ratio of reactants was obtained by changing the ratio of reactants while other reaction parameters were kept constant. For synthesis of MFHAs the optimum mol ratio of N-MHA/palm kernel oil = 6/1 and the highest conversion was 77.8%, for synthesis of IPFHAs the optimum mol ratio of N-IPHA/palm kernel oil = 7/1 and the highest conversion was 65.4% and for synthesis of BFHAs the optimum mol ratio of N-BHA/palm kernel oil = 7/1 and the highest conversion was 61.7%.
Subspace exclusion zones for damage localization
Bernal, Dionisio; Ulriksen, Martin Dalgaard
2018-01-01
, this is exploited in the context of structural damage localization to cast the Subspace Exclusion Zone (SEZ) scheme, which locates damage by reconstructing the captured field quantity shifts from analytical subspaces indexed by postulated boundaries, the so-called exclusion zones (EZs), in a model of the structure...
Kernel versions of some orthogonal transformations
Nielsen, Allan Aasbjerg
Kernel versions of orthogonal transformations such as principal components are based on a dual formulation also termed Q-mode analysis in which the data enter into the analysis via inner products in the Gram matrix only. In the kernel version the inner products of the original data are replaced...... by inner products between nonlinear mappings into higher dimensional feature space. Via kernel substitution also known as the kernel trick these inner products between the mappings are in turn replaced by a kernel function and all quantities needed in the analysis are expressed in terms of this kernel...... function. This means that we need not know the nonlinear mappings explicitly. Kernel principal component analysis (PCA) and kernel minimum noise fraction (MNF) analyses handle nonlinearities by implicitly transforming data into high (even infinite) dimensional feature space via the kernel function...
An Approximate Approach to Automatic Kernel Selection.
Ding, Lizhong; Liao, Shizhong
2016-02-02
Kernel selection is a fundamental problem of kernel-based learning algorithms. In this paper, we propose an approximate approach to automatic kernel selection for regression from the perspective of kernel matrix approximation. We first introduce multilevel circulant matrices into automatic kernel selection, and develop two approximate kernel selection algorithms by exploiting the computational virtues of multilevel circulant matrices. The complexity of the proposed algorithms is quasi-linear in the number of data points. Then, we prove an approximation error bound to measure the effect of the approximation in kernel matrices by multilevel circulant matrices on the hypothesis and further show that the approximate hypothesis produced with multilevel circulant matrices converges to the accurate hypothesis produced with kernel matrices. Experimental evaluations on benchmark datasets demonstrate the effectiveness of approximate kernel selection.
Model Selection in Kernel Ridge Regression
Exterkate, Peter
Kernel ridge regression is gaining popularity as a data-rich nonlinear forecasting tool, which is applicable in many different contexts. This paper investigates the influence of the choice of kernel and the setting of tuning parameters on forecast accuracy. We review several popular kernels......, including polynomial kernels, the Gaussian kernel, and the Sinc kernel. We interpret the latter two kernels in terms of their smoothing properties, and we relate the tuning parameters associated to all these kernels to smoothness measures of the prediction function and to the signal-to-noise ratio. Based...... on these interpretations, we provide guidelines for selecting the tuning parameters from small grids using cross-validation. A Monte Carlo study confirms the practical usefulness of these rules of thumb. Finally, the flexible and smooth functional forms provided by the Gaussian and Sinc kernels makes them widely...
Adeli, Ehsan; Wu, Guorong; Saghafi, Behrouz; An, Le; Shi, Feng; Shen, Dinggang
2017-01-01
Feature selection methods usually select the most compact and relevant set of features based on their contribution to a linear regression model. Thus, these features might not be the best for a non-linear classifier. This is especially crucial for the tasks, in which the performance is heavily dependent on the feature selection techniques, like the diagnosis of neurodegenerative diseases. Parkinson’s disease (PD) is one of the most common neurodegenerative disorders, which progresses slowly while affects the quality of life dramatically. In this paper, we use the data acquired from multi-modal neuroimaging data to diagnose PD by investigating the brain regions, known to be affected at the early stages. We propose a joint kernel-based feature selection and classification framework. Unlike conventional feature selection techniques that select features based on their performance in the original input feature space, we select features that best benefit the classification scheme in the kernel space. We further propose kernel functions, specifically designed for our non-negative feature types. We use MRI and SPECT data of 538 subjects from the PPMI database, and obtain a diagnosis accuracy of 97.5%, which outperforms all baseline and state-of-the-art methods.
Cheong, Kwang-Ho; Suh, Tae-Suk; Lee, Hyoung-Koo; Choe, Bo-Young [The Catholic Univ. of Korea, Seoul (Korea, Republic of); Kim, Hoi-Nam; Yoon, Sei-Chul [Kangnam St. Mary' s Hospital, Seoul (Korea, Republic of)
2002-07-01
Accurate dose calculation in radiation treatment planning is most important for successful treatment. Since human body is composed of various materials and not an ideal shape, it is not easy to calculate the accurate effective dose in the patients. Many methods have been proposed to solve inhomogeneity and surface contour problems. Monte Carlo simulations are regarded as the most accurate method, but it is not appropriate for routine planning because it takes so much time. Pencil beam kernel based convolution/superposition methods were also proposed to correct those effects. Nowadays, many commercial treatment planning systems have adopted this algorithm as a dose calculation engine. The purpose of this study is to verify the accuracy of the dose calculated from pencil beam kernel based treatment planning system comparing to Monte Carlo simulations and measurements especially in inhomogeneous region. Home-made inhomogeneous phantom, Helax-TMS ver. 6.0 and Monte Carlo code BEAMnrc and DOSXYZnrc were used in this study. In homogeneous media, the accuracy was acceptable but in inhomogeneous media, the errors were more significant. However in general clinical situation, pencil beam kernel based convolution algorithm is thought to be a valuable tool to calculate the dose.
Adeli, Ehsan; Wu, Guorong; Saghafi, Behrouz; An, Le; Shi, Feng; Shen, Dinggang
2017-01-01
Feature selection methods usually select the most compact and relevant set of features based on their contribution to a linear regression model. Thus, these features might not be the best for a non-linear classifier. This is especially crucial for the tasks, in which the performance is heavily dependent on the feature selection techniques, like the diagnosis of neurodegenerative diseases. Parkinson’s disease (PD) is one of the most common neurodegenerative disorders, which progresses slowly while affects the quality of life dramatically. In this paper, we use the data acquired from multi-modal neuroimaging data to diagnose PD by investigating the brain regions, known to be affected at the early stages. We propose a joint kernel-based feature selection and classification framework. Unlike conventional feature selection techniques that select features based on their performance in the original input feature space, we select features that best benefit the classification scheme in the kernel space. We further propose kernel functions, specifically designed for our non-negative feature types. We use MRI and SPECT data of 538 subjects from the PPMI database, and obtain a diagnosis accuracy of 97.5%, which outperforms all baseline and state-of-the-art methods. PMID:28120883
Ahmed Elazab
2015-01-01
Full Text Available An adaptively regularized kernel-based fuzzy C-means clustering framework is proposed for segmentation of brain magnetic resonance images. The framework can be in the form of three algorithms for the local average grayscale being replaced by the grayscale of the average filter, median filter, and devised weighted images, respectively. The algorithms employ the heterogeneity of grayscales in the neighborhood and exploit this measure for local contextual information and replace the standard Euclidean distance with Gaussian radial basis kernel functions. The main advantages are adaptiveness to local context, enhanced robustness to preserve image details, independence of clustering parameters, and decreased computational costs. The algorithms have been validated against both synthetic and clinical magnetic resonance images with different types and levels of noises and compared with 6 recent soft clustering algorithms. Experimental results show that the proposed algorithms are superior in preserving image details and segmentation accuracy while maintaining a low computational complexity.
An OSKit-Based Implementation of Least Privilege Separation Kernel Memory Partitioning
2007-06-01
Fiasco is developed at TU Dresden. It is compatible with the x86 L4 microkernel . It is a real-time preemptive kernel written in C++. Fiasco’s...adherence to the specification of the L4 microkernel makes it a desirable choice. The downside is Fiasco is too specialized for what is needed by the
DETECTION OF CHANGES OF THE SYSTEM TECHNICAL STATE USING STOCHASTIC SUBSPACE OBSERVATION METHOD
Andrzej Puchalski
2014-03-01
Full Text Available System diagnostics based on vibroacoustics signals, carried out by means of stochastic subspace methods was undertaken in the hereby paper. Subspace methods are the ones based on numerical linear algebra tools. The considered solutions belong to diagnostic methods according to data, leading to the generation of residuals allowing failure recognition of elements and assemblies in machines and devices. The algorithm of diagnostics according to the subspace observation method was applied – in the paper – for the estimation of the valve system of the spark ignition engine.
A kernel version of multivariate alteration detection
Nielsen, Allan Aasbjerg; Vestergaard, Jacob Schack
2013-01-01
Based on the established methods kernel canonical correlation analysis and multivariate alteration detection we introduce a kernel version of multivariate alteration detection. A case study with SPOT HRV data shows that the kMAD variates focus on extreme change observations.......Based on the established methods kernel canonical correlation analysis and multivariate alteration detection we introduce a kernel version of multivariate alteration detection. A case study with SPOT HRV data shows that the kMAD variates focus on extreme change observations....
Droplet Image Super Resolution Based on Sparse Representation and Kernel Regression
Zou, Zhenzhen; Luo, Xinghong; Yu, Qiang
2018-05-01
Microgravity and containerless conditions, which are produced via electrostatic levitation combined with a drop tube, are important when studying the intrinsic properties of new metastable materials. Generally, temperature and image sensors can be used to measure the changes of sample temperature, morphology and volume. Then, the specific heat, surface tension, viscosity changes and sample density can be obtained. Considering that the falling speed of the material sample droplet is approximately 31.3 m/s when it reaches the bottom of a 50-meter-high drop tube, a high-speed camera with a collection rate of up to 106 frames/s is required to image the falling droplet. However, at the high-speed mode, very few pixels, approximately 48-120, will be obtained in each exposure time, which results in low image quality. Super-resolution image reconstruction is an algorithm that provides finer details than the sampling grid of a given imaging device by increasing the number of pixels per unit area in the image. In this work, we demonstrate the application of single image-resolution reconstruction in the microgravity and electrostatic levitation for the first time. Here, using the image super-resolution method based on sparse representation, a low-resolution droplet image can be reconstructed. Employed Yang's related dictionary model, high- and low-resolution image patches were combined with dictionary training, and high- and low-resolution-related dictionaries were obtained. The online double-sparse dictionary training algorithm was used in the study of related dictionaries and overcome the shortcomings of the traditional training algorithm with small image patch. During the stage of image reconstruction, the algorithm of kernel regression is added, which effectively overcomes the shortcomings of the Yang image's edge blurs.
Code subspaces for LLM geometries
Berenstein, David; Miller, Alexandra
2018-03-01
We consider effective field theory around classical background geometries with a gauge theory dual, specifically those in the class of LLM geometries. These are dual to half-BPS states of N= 4 SYM. We find that the language of code subspaces is natural for discussing the set of nearby states, which are built by acting with effective fields on these backgrounds. This work extends our previous work by going beyond the strict infinite N limit. We further discuss how one can extract the topology of the state beyond N→∞ and find that, as before, uncertainty and entanglement entropy calculations provide a useful tool to do so. Finally, we discuss obstructions to writing down a globally defined metric operator. We find that the answer depends on the choice of reference state that one starts with. Therefore, within this setup, there is ambiguity in trying to write an operator that describes the metric globally.
Schumacher, F.; Friederich, W.
2015-12-01
We present the modularized software package ASKI which is a flexible and extendable toolbox for seismic full waveform inversion (FWI) as well as sensitivity or resolution analysis operating on the sensitivity matrix. It utilizes established wave propagation codes for solving the forward problem and offers an alternative to the monolithic, unflexible and hard-to-modify codes that have typically been written for solving inverse problems. It is available under the GPL at www.rub.de/aski. The Gauss-Newton FWI method for 3D-heterogeneous elastic earth models is based on waveform sensitivity kernels and can be applied to inverse problems at various spatial scales in both Cartesian and spherical geometries. The kernels are derived in the frequency domain from Born scattering theory as the Fréchet derivatives of linearized full waveform data functionals, quantifying the influence of elastic earth model parameters on the particular waveform data values. As an important innovation, we keep two independent spatial descriptions of the earth model - one for solving the forward problem and one representing the inverted model updates. Thereby we account for the independent needs of spatial model resolution of forward and inverse problem, respectively. Due to pre-integration of the kernels over the (in general much coarser) inversion grid, storage requirements for the sensitivity kernels are dramatically reduced.ASKI can be flexibly extended to other forward codes by providing it with specific interface routines that contain knowledge about forward code-specific file formats and auxiliary information provided by the new forward code. In order to sustain flexibility, the ASKI tools must communicate via file output/input, thus large storage capacities need to be accessible in a convenient way. Storing the complete sensitivity matrix to file, however, permits the scientist full manual control over each step in a customized procedure of sensitivity/resolution analysis and full
Mailys Lopes
2017-07-01
Full Text Available This paper deals with the classification of grasslands using high resolution satellite image time series. Grasslands considered in this work are semi-natural elements in fragmented landscapes, i.e., they are heterogeneous and small elements. The first contribution of this study is to account for grassland heterogeneity while working at the object level by modeling its pixels distributions by a Gaussian distribution. To measure the similarity between two grasslands, a new kernel is proposed as a second contribution: the α -Gaussian mean kernel. It allows one to weight the influence of the covariance matrix when comparing two Gaussian distributions. This kernel is introduced in support vector machines for the supervised classification of grasslands from southwest France. A dense intra-annual multispectral time series of the Formosat-2 satellite is used for the classification of grasslands’ management practices, while an inter-annual NDVI time series of Formosat-2 is used for old and young grasslands’ discrimination. Results are compared to other existing pixel- and object-based approaches in terms of classification accuracy and processing time. The proposed method is shown to be a good compromise between processing speed and classification accuracy. It can adapt to the classification constraints, and it encompasses several similarity measures known in the literature. It is appropriate for the classification of small and heterogeneous objects such as grasslands.
Du, Genyuan; Tian, Shengli; Qiu, Yingyu; Xu, Chunyan
2016-01-01
This paper presents an effective and efficient kernel approach to recognize image set which is represented as a point on extended Grassmannian manifold. Several recent studies focus on the applicability of discriminant analysis on Grassmannian manifold and suffer from not obtaining the inherent nonlinear structure of the data itself. Therefore, we propose an extension of Grassmannian manifold to address this issue. Instead of using a linear data embedding with PCA, we develop a non-linear data embedding of such manifold using kernel PCA. This paper mainly consider three folds: 1) introduce a non-linear data embedding of extended Grassmannian manifold, 2) derive a distance metric of Grassmannian manifold, 3) develop an effective and efficient Grassmannian kernel for SVM classification. The extended Grassmannian manifold naturally arises in the application to recognition based on image set, such as face and object recognition. Experiments on several standard databases show better classification accuracy. Furthermore, experimental results indicate that our proposed approach significantly reduces time complexity in comparison to graph embedding discriminant analysis.
Barndorff-Nielsen, Ole Eiler; Hansen, P. Reinhard; Lunde, Asger
2009-01-01
and find a remarkable level of agreement. We identify some features of the high-frequency data, which are challenging for realized kernels. They are when there are local trends in the data, over periods of around 10 minutes, where the prices and quotes are driven up or down. These can be associated......Realized kernels use high-frequency data to estimate daily volatility of individual stock prices. They can be applied to either trade or quote data. Here we provide the details of how we suggest implementing them in practice. We compare the estimates based on trade and quote data for the same stock...
Geometric subspace updates with applications to online adaptive nonlinear model reduction
Zimmermann, Ralf; Peherstorfer, Benjamin; Willcox, Karen
2018-01-01
In many scientific applications, including model reduction and image processing, subspaces are used as ansatz spaces for the low-dimensional approximation and reconstruction of the state vectors of interest. We introduce a procedure for adapting an existing subspace based on information from...... Estimation (GROUSE). We establish for GROUSE a closed-form expression for the residual function along the geodesic descent direction. Specific applications of subspace adaptation are discussed in the context of image processing and model reduction of nonlinear partial differential equation systems....
ASCS online fault detection and isolation based on an improved MPCA
Peng, Jianxin; Liu, Haiou; Hu, Yuhui; Xi, Junqiang; Chen, Huiyan
2014-09-01
Multi-way principal component analysis (MPCA) has received considerable attention and been widely used in process monitoring. A traditional MPCA algorithm unfolds multiple batches of historical data into a two-dimensional matrix and cut the matrix along the time axis to form subspaces. However, low efficiency of subspaces and difficult fault isolation are the common disadvantages for the principal component model. This paper presents a new subspace construction method based on kernel density estimation function that can effectively reduce the storage amount of the subspace information. The MPCA model and the knowledge base are built based on the new subspace. Then, fault detection and isolation with the squared prediction error (SPE) statistic and the Hotelling ( T 2) statistic are also realized in process monitoring. When a fault occurs, fault isolation based on the SPE statistic is achieved by residual contribution analysis of different variables. For fault isolation of subspace based on the T 2 statistic, the relationship between the statistic indicator and state variables is constructed, and the constraint conditions are presented to check the validity of fault isolation. Then, to improve the robustness of fault isolation to unexpected disturbances, the statistic method is adopted to set the relation between single subspace and multiple subspaces to increase the corrective rate of fault isolation. Finally fault detection and isolation based on the improved MPCA is used to monitor the automatic shift control system (ASCS) to prove the correctness and effectiveness of the algorithm. The research proposes a new subspace construction method to reduce the required storage capacity and to prove the robustness of the principal component model, and sets the relationship between the state variables and fault detection indicators for fault isolation.
Veto-Consensus Multiple Kernel Learning
Zhou, Y.; Hu, N.; Spanos, C.J.
2016-01-01
We propose Veto-Consensus Multiple Kernel Learning (VCMKL), a novel way of combining multiple kernels such that one class of samples is described by the logical intersection (consensus) of base kernelized decision rules, whereas the other classes by the union (veto) of their complements. The
A Spectral-Texture Kernel-Based Classification Method for Hyperspectral Images
Yi Wang
2016-11-01
Full Text Available Classification of hyperspectral images always suffers from high dimensionality and very limited labeled samples. Recently, the spectral-spatial classification has attracted considerable attention and can achieve higher classification accuracy and smoother classification maps. In this paper, a novel spectral-spatial classification method for hyperspectral images by using kernel methods is investigated. For a given hyperspectral image, the principle component analysis (PCA transform is first performed. Then, the first principle component of the input image is segmented into non-overlapping homogeneous regions by using the entropy rate superpixel (ERS algorithm. Next, the local spectral histogram model is applied to each homogeneous region to obtain the corresponding texture features. Because this step is performed within each homogenous region, instead of within a fixed-size image window, the obtained local texture features in the image are more accurate, which can effectively benefit the improvement of classification accuracy. In the following step, a contextual spectral-texture kernel is constructed by combining spectral information in the image and the extracted texture information using the linearity property of the kernel methods. Finally, the classification map is achieved by the support vector machines (SVM classifier using the proposed spectral-texture kernel. Experiments on two benchmark airborne hyperspectral datasets demonstrate that our method can effectively improve classification accuracies, even though only a very limited training sample is available. Specifically, our method can achieve from 8.26% to 15.1% higher in terms of overall accuracy than the traditional SVM classifier. The performance of our method was further compared to several state-of-the-art classification methods of hyperspectral images using objective quantitative measures and a visual qualitative evaluation.
Performance of fly ash based geopolymer incorporating palm kernel shell for lightweight concrete
Razak, Rafiza Abd; Abdullah, Mohd Mustafa Al Bakri; Yahya, Zarina; Jian, Ang Zhi; Nasri, Armia
2017-09-01
A concrete which cement is totally replaced by source material such as fly ash and activated by highly alkaline solutions is known as geopolymer concrete. Fly ash is the most common source material for geopolymer because it is a by-product material, so it can get easily from all around the world. An investigation has been carried out to select the most suitable ingredients of geopolymer concrete so that the geopolymer concrete can achieve the desire compressive strength. The samples were prepared to determine the suitable percentage of palm kernel shell used in geopolymer concrete and cured for 7 days in oven. After that, other samples were prepared by using the suitable percentage of palm kernel shell and cured for 3, 14, 21 and 28 days in oven. The control sample consisting of ordinary Portland cement and palm kernel shell and cured for 28 days were prepared too. The NaOH concentration of 12M, ratio Na2SiO3 to NaOH of 2.5, ratio fly ash to alkaline activator solution of 2.0 and ratio water to geopolymer of 0.35 were fixed throughout the research. The density obtained for the samples were 1.78 kg/m3, water absorption of 20.41% and the compressive strength of 14.20 MPa. The compressive strength of geopolymer concrete is still acceptable as lightweight concrete although the compressive strength is lower than OPC concrete. Therefore, the proposed method by using fly ash mixed with 10% of palm kernel shell can be used to design geopolymer concrete.
Jiang, He; Dong, Yao
2017-01-01
Highlights: • The structured variable selection in Kernel SVM is implemented using two ways. • The two-way interaction model is considered to enforce Heredity Principle. • SVMIC is used to select the kernel parameter in proposed approaches. • Simple and fast computations algorithms are derived. - Abstract: Various applications of forecasting effective global horizontal irradiance play increasingly vital role in grid-connected photovoltaic installations, but suffer from forecasting inaccuracy and prohibitively expensive computational cost. Although Support Vector Machine (SVM) is one of the most powerful forecasting approaches, it does not provide an interpretable model. This motivates penalized variable selection methods to be introduced to SVM to select important variables. However, in some forecasting problems, there are some underlying logic or hierarchical structure such as heredity principle among the variables. Penalized Kernel SVM approaches do not take heredity principles into consideration when enforcing sparsity. This paper investigates structural variable selection in Kernel SVM based approach which pursues heredity principle and sparsity simultaneously. To achieve heredity principle, both optimization and procedure based structural variable selection approaches are studied in the Kernel SVM. Computationally, we derive fast and simple-to-implement algorithms to perform structural variable selection and solar irradiance forecasting. Furthermore, Support Vector Machines Information Criterion is utilized to select the kernel parameters to guarantee the model consistency. Real data experiments directly reveal that our proposed KSVM-SVS based approach following heredity principle delivers superior performances in terms of forecasting accuracy comparing with other competitors.
Model selection in kernel ridge regression
Exterkate, Peter
2013-01-01
Kernel ridge regression is a technique to perform ridge regression with a potentially infinite number of nonlinear transformations of the independent variables as regressors. This method is gaining popularity as a data-rich nonlinear forecasting tool, which is applicable in many different contexts....... The influence of the choice of kernel and the setting of tuning parameters on forecast accuracy is investigated. Several popular kernels are reviewed, including polynomial kernels, the Gaussian kernel, and the Sinc kernel. The latter two kernels are interpreted in terms of their smoothing properties......, and the tuning parameters associated to all these kernels are related to smoothness measures of the prediction function and to the signal-to-noise ratio. Based on these interpretations, guidelines are provided for selecting the tuning parameters from small grids using cross-validation. A Monte Carlo study...
Wang, Jim Jing-Yan
2014-12-01
Trypanosma brucei (T. Brucei) is an important pathogen agent of African trypanosomiasis. The flagellum is an essential and multifunctional organelle of T. Brucei, thus it is very important to recognize the flagellar proteins from T. Brucei proteins for the purposes of both biological research and drug design. In this paper, we investigate computationally recognizing flagellar proteins in T. Brucei by pattern recognition methods. It is argued that an optimal decision function can be obtained as the difference of probability functions of flagella protein and the non-flagellar protein for the purpose of flagella protein recognition. We propose to learn a multi-kernel classification function to approximate this optimal decision function, by minimizing the information loss of such approximation which is measured by the Kull back-Leibler (KL) divergence. An iterative multi-kernel classifier learning algorithm is developed to minimize the KL divergence for the problem of T. Brucei flagella protein recognition, experiments show its advantage over other T. Brucei flagellar protein recognition and multi-kernel learning methods. © 2014 IEEE.
A Fisher Kernel Approach for Multiple Instance Based Object Retrieval in Video Surveillance
MIRONICA, I.
2015-11-01
Full Text Available This paper presents an automated surveillance system that exploits the Fisher Kernel representation in the context of multiple-instance object retrieval task. The proposed algorithm has the main purpose of tracking a list of persons in several video sources, using only few training examples. In the first step, the Fisher Kernel representation describes a set of features as the derivative with respect to the log-likelihood of the generative probability distribution that models the feature distribution. Then, we learn the generative probability distribution over all features extracted from a reduced set of relevant frames. The proposed approach shows significant improvements and we demonstrate that Fisher kernels are well suited for this task. We demonstrate the generality of our approach in terms of features by conducting an extensive evaluation with a broad range of keypoints features. Also, we evaluate our method on two standard video surveillance datasets attaining superior results comparing to state-of-the-art object recognition algorithms.
Xiong Luo
2016-07-01
Full Text Available With the recent emergence of wireless sensor networks (WSNs in the cloud computing environment, it is now possible to monitor and gather physical information via lots of sensor nodes to meet the requirements of cloud services. Generally, those sensor nodes collect data and send data to sink node where end-users can query all the information and achieve cloud applications. Currently, one of the main disadvantages in the sensor nodes is that they are with limited physical performance relating to less memory for storage and less source of power. Therefore, in order to avoid such limitation, it is necessary to develop an efficient data prediction method in WSN. To serve this purpose, by reducing the redundant data transmission between sensor nodes and sink node while maintaining the required acceptable errors, this article proposes an entropy-based learning scheme for data prediction through the use of kernel least mean square (KLMS algorithm. The proposed scheme called E-KLMS develops a mechanism to maintain the predicted data synchronous at both sides. Specifically, the kernel-based method is able to adjust the coefficients adaptively in accordance with every input, which will achieve a better performance with smaller prediction errors, while employing information entropy to remove these data which may cause relatively large errors. E-KLMS can effectively solve the tradeoff problem between prediction accuracy and computational efforts while greatly simplifying the training structure compared with some other data prediction approaches. What’s more, the kernel-based method and entropy technique could ensure the prediction effect by both improving the accuracy and reducing errors. Experiments with some real data sets have been carried out to validate the efficiency and effectiveness of E-KLMS learning scheme, and the experiment results show advantages of the our method in prediction accuracy and computational time.
Integrated Phoneme Subspace Method for Speech Feature Extraction
Park Hyunsin
2009-01-01
Full Text Available Speech feature extraction has been a key focus in robust speech recognition research. In this work, we discuss data-driven linear feature transformations applied to feature vectors in the logarithmic mel-frequency filter bank domain. Transformations are based on principal component analysis (PCA, independent component analysis (ICA, and linear discriminant analysis (LDA. Furthermore, this paper introduces a new feature extraction technique that collects the correlation information among phoneme subspaces and reconstructs feature space for representing phonemic information efficiently. The proposed speech feature vector is generated by projecting an observed vector onto an integrated phoneme subspace (IPS based on PCA or ICA. The performance of the new feature was evaluated for isolated word speech recognition. The proposed method provided higher recognition accuracy than conventional methods in clean and reverberant environments.
Qiang Shang
2016-08-01
Full Text Available Short-term traffic flow prediction is an important part of intelligent transportation systems research and applications. For further improving the accuracy of short-time traffic flow prediction, a novel hybrid prediction model (multivariate phase space reconstruction–combined kernel function-least squares support vector machine based on multivariate phase space reconstruction and combined kernel function-least squares support vector machine is proposed. The C-C method is used to determine the optimal time delay and the optimal embedding dimension of traffic variables’ (flow, speed, and occupancy time series for phase space reconstruction. The G-P method is selected to calculate the correlation dimension of attractor which is an important index for judging chaotic characteristics of the traffic variables’ series. The optimal input form of combined kernel function-least squares support vector machine model is determined by multivariate phase space reconstruction, and the model’s parameters are optimized by particle swarm optimization algorithm. Finally, case validation is carried out using the measured data of an expressway in Xiamen, China. The experimental results suggest that the new proposed model yields better predictions compared with similar models (combined kernel function-least squares support vector machine, multivariate phase space reconstruction–generalized kernel function-least squares support vector machine, and phase space reconstruction–combined kernel function-least squares support vector machine, which indicates that the new proposed model exhibits stronger prediction ability and robustness.
A new method by steering kernel-based Richardson–Lucy algorithm for neutron imaging restoration
Qiao, Shuang; Wang, Qiao; Sun, Jia-ning; Huang, Ji-peng
2014-01-01
Motivated by industrial applications, neutron radiography has become a powerful tool for non-destructive investigation techniques. However, resulted from a combined effect of neutron flux, collimated beam, limited spatial resolution of detector and scattering, etc., the images made with neutrons are degraded severely by blur and noise. For dealing with it, by integrating steering kernel regression into Richardson–Lucy approach, we present a novel restoration method in this paper, which is capable of suppressing noise while restoring details of the blurred imaging result efficiently. Experimental results show that compared with the other methods, the proposed method can improve the restoration quality both visually and quantitatively
Zhigao Zeng
2016-01-01
Full Text Available This paper proposes a novel algorithm to solve the challenging problem of classifying error-diffused halftone images. We firstly design the class feature matrices, after extracting the image patches according to their statistics characteristics, to classify the error-diffused halftone images. Then, the spectral regression kernel discriminant analysis is used for feature dimension reduction. The error-diffused halftone images are finally classified using an idea similar to the nearest centroids classifier. As demonstrated by the experimental results, our method is fast and can achieve a high classification accuracy rate with an added benefit of robustness in tackling noise.
Valentini, Giorgio; Paccanaro, Alberto; Caniza, Horacio; Romero, Alfonso E; Re, Matteo
2014-06-01
In the context of "network medicine", gene prioritization methods represent one of the main tools to discover candidate disease genes by exploiting the large amount of data covering different types of functional relationships between genes. Several works proposed to integrate multiple sources of data to improve disease gene prioritization, but to our knowledge no systematic studies focused on the quantitative evaluation of the impact of network integration on gene prioritization. In this paper, we aim at providing an extensive analysis of gene-disease associations not limited to genetic disorders, and a systematic comparison of different network integration methods for gene prioritization. We collected nine different functional networks representing different functional relationships between genes, and we combined them through both unweighted and weighted network integration methods. We then prioritized genes with respect to each of the considered 708 medical subject headings (MeSH) diseases by applying classical guilt-by-association, random walk and random walk with restart algorithms, and the recently proposed kernelized score functions. The results obtained with classical random walk algorithms and the best single network achieved an average area under the curve (AUC) across the 708 MeSH diseases of about 0.82, while kernelized score functions and network integration boosted the average AUC to about 0.89. Weighted integration, by exploiting the different "informativeness" embedded in different functional networks, outperforms unweighted integration at 0.01 significance level, according to the Wilcoxon signed rank sum test. For each MeSH disease we provide the top-ranked unannotated candidate genes, available for further bio-medical investigation. Network integration is necessary to boost the performances of gene prioritization methods. Moreover the methods based on kernelized score functions can further enhance disease gene ranking results, by adopting both
Valentini, Giorgio; Paccanaro, Alberto; Caniza, Horacio; Romero, Alfonso E.; Re, Matteo
2014-01-01
Objective In the context of “network medicine”, gene prioritization methods represent one of the main tools to discover candidate disease genes by exploiting the large amount of data covering different types of functional relationships between genes. Several works proposed to integrate multiple sources of data to improve disease gene prioritization, but to our knowledge no systematic studies focused on the quantitative evaluation of the impact of network integration on gene prioritization. In this paper, we aim at providing an extensive analysis of gene-disease associations not limited to genetic disorders, and a systematic comparison of different network integration methods for gene prioritization. Materials and methods We collected nine different functional networks representing different functional relationships between genes, and we combined them through both unweighted and weighted network integration methods. We then prioritized genes with respect to each of the considered 708 medical subject headings (MeSH) diseases by applying classical guilt-by-association, random walk and random walk with restart algorithms, and the recently proposed kernelized score functions. Results The results obtained with classical random walk algorithms and the best single network achieved an average area under the curve (AUC) across the 708 MeSH diseases of about 0.82, while kernelized score functions and network integration boosted the average AUC to about 0.89. Weighted integration, by exploiting the different “informativeness” embedded in different functional networks, outperforms unweighted integration at 0.01 significance level, according to the Wilcoxon signed rank sum test. For each MeSH disease we provide the top-ranked unannotated candidate genes, available for further bio-medical investigation. Conclusions Network integration is necessary to boost the performances of gene prioritization methods. Moreover the methods based on kernelized score functions can further
Different structures on subspaces of OsckM
Čomić Irena
2013-01-01
Full Text Available The geometry of OsckM spaces was introduced by R. Miron and Gh. Atanasiu in [6] and [7]. The theory of these spaces was developed by R. Miron and his cooperators from Romania, Japan and other countries in several books and many papers. Only some of them are mentioned in references. Here we recall the construction of adapted bases in T(OsckM and T*(OsckM, which are comprehensive with the J structure. The theory of two complementary family of subspaces is presented as it was done in [2] and [4]. The operators J,J, θ,θ, p, p* are introduced in the ambient space and subspaces. Some new relations between them are established. The action of these operators on Liouville vector fields are examined.
Chmiel, Malgorzata; Roux, Philippe; Herrmann, Philippe; Rondeleux, Baptiste; Wathelet, Marc
2018-05-01
We investigated the construction of diffraction kernels for surface waves using two-point convolution and/or correlation from land active seismic data recorded in the context of exploration geophysics. The high density of controlled sources and receivers, combined with the application of the reciprocity principle, allows us to retrieve two-dimensional phase-oscillation diffraction kernels (DKs) of surface waves between any two source or receiver points in the medium at each frequency (up to 15 Hz, at least). These DKs are purely data-based as no model calculations and no synthetic data are needed. They naturally emerge from the interference patterns of the recorded wavefields projected on the dense array of sources and/or receivers. The DKs are used to obtain multi-mode dispersion relations of Rayleigh waves, from which near-surface shear velocity can be extracted. Using convolution versus correlation with a grid of active sources is an important step in understanding the physics of the retrieval of surface wave Green's functions. This provides the foundation for future studies based on noise sources or active sources with a sparse spatial distribution.
Amir Hossein Mirzabe
2017-12-01
Full Text Available In this study, a method based on digital image processing was employed to investigate effect of the moisture content on gravimetrical and frictional properties of the cucumber seeds and kernels. This research indicated that the application of visual machines and the image processing could be a rapid method that enables accurate measurement of the dimensional parameters of two varieties of this product meticulously. The length, width, and thickness of the seeds of Rashid variety ranged from 6.40 to 9.07, 2.91 to 4.21 and 0.65 to1.50 mm, respectively; the corresponding value of Negin variety ranged from 6.89 to 9.07, 2.49 to 4.21, and 0.69 to 1.68 mm, respectively. Kernel and shell ratios of the Rashid variety were found to be 69.811 and 30.189%, respectively; the corresponding values of the Negin variety were found to be 72.727 and 27.273%, respectively. Despite the bulk densities of the two varieties of cucumber seeds and kernels that decreased with the moisture content, true density and porosity of the two varieties of cucumber seeds and kernels increased. As the moisture content increased from 5.04 to 21.12% (d.b, the angle of the static friction of the seeds of Negin variety increased from 30.39° to 37.18°, 28.83° to 34.99°, 24.37° to 30.04°, and 15.77° to 19.57° for wood, rubber, iron, and galvanized, respectively,; the corresponding value of the Negin variety increased from 21.12° to 27.16°, 24.43° to 31.41°, 19.90° to 25.59° and 14.09° to 18.12° as the moisture content increased from 5.02 to 21.02% (d.b. Keywords: Image processing, Physical and mechanical properties, Cucumber
Active Subspaces of Airfoil Shape Parameterizations
Grey, Zachary J.; Constantine, Paul G.
2018-05-01
Design and optimization benefit from understanding the dependence of a quantity of interest (e.g., a design objective or constraint function) on the design variables. A low-dimensional active subspace, when present, identifies important directions in the space of design variables; perturbing a design along the active subspace associated with a particular quantity of interest changes that quantity more, on average, than perturbing the design orthogonally to the active subspace. This low-dimensional structure provides insights that characterize the dependence of quantities of interest on design variables. Airfoil design in a transonic flow field with a parameterized geometry is a popular test problem for design methodologies. We examine two particular airfoil shape parameterizations, PARSEC and CST, and study the active subspaces present in two common design quantities of interest, transonic lift and drag coefficients, under each shape parameterization. We mathematically relate the two parameterizations with a common polynomial series. The active subspaces enable low-dimensional approximations of lift and drag that relate to physical airfoil properties. In particular, we obtain and interpret a two-dimensional approximation of both transonic lift and drag, and we show how these approximation inform a multi-objective design problem.
Starke, G. [Universitaet Karlsruhe (Germany)
1994-12-31
For nonselfadjoint elliptic boundary value problems which are preconditioned by a substructuring method, i.e., nonoverlapping domain decomposition, the author introduces and studies the concept of subspace orthogonalization. In subspace orthogonalization variants of Krylov methods the computation of inner products and vector updates, and the storage of basis elements is restricted to a (presumably small) subspace, in this case the edge and vertex unknowns with respect to the partitioning into subdomains. The author investigates subspace orthogonalization for two specific iterative algorithms, GMRES and the full orthogonalization method (FOM). This is intended to eliminate certain drawbacks of the Arnoldi-based Krylov subspace methods mentioned above. Above all, the length of the Arnoldi recurrences grows linearly with the iteration index which is therefore restricted to the number of basis elements that can be held in memory. Restarts become necessary and this often results in much slower convergence. The subspace orthogonalization methods, in contrast, require the storage of only the edge and vertex unknowns of each basis element which means that one can iterate much longer before restarts become necessary. Moreover, the computation of inner products is also restricted to the edge and vertex points which avoids the disturbance of the computational flow associated with the solution of subdomain problems. The author views subspace orthogonalization as an alternative to restarting or truncating Krylov subspace methods for nonsymmetric linear systems of equations. Instead of shortening the recurrences, one restricts them to a subset of the unknowns which has to be carefully chosen in order to be able to extend this partial solution to the entire space. The author discusses the convergence properties of these iteration schemes and its advantages compared to restarted or truncated versions of Krylov methods applied to the full preconditioned system.
Gong, Kuang; Cheng-Liao, Jinxiu; Wang, Guobao; Chen, Kevin T; Catana, Ciprian; Qi, Jinyi
2018-04-01
Positron emission tomography (PET) is a functional imaging modality widely used in oncology, cardiology, and neuroscience. It is highly sensitive, but suffers from relatively poor spatial resolution, as compared with anatomical imaging modalities, such as magnetic resonance imaging (MRI). With the recent development of combined PET/MR systems, we can improve the PET image quality by incorporating MR information into image reconstruction. Previously, kernel learning has been successfully embedded into static and dynamic PET image reconstruction using either PET temporal or MRI information. Here, we combine both PET temporal and MRI information adaptively to improve the quality of direct Patlak reconstruction. We examined different approaches to combine the PET and MRI information in kernel learning to address the issue of potential mismatches between MRI and PET signals. Computer simulations and hybrid real-patient data acquired on a simultaneous PET/MR scanner were used to evaluate the proposed methods. Results show that the method that combines PET temporal information and MRI spatial information adaptively based on the structure similarity index has the best performance in terms of noise reduction and resolution improvement.
Huang, Xia; Li, Chunqiang; Xiao, Chuan; Sun, Wenqing; Qian, Wei
2017-03-01
The temporal focusing two-photon microscope (TFM) is developed to perform depth resolved wide field fluorescence imaging by capturing frames sequentially. However, due to strong nonignorable noises and diffraction rings surrounding particles, further researches are extremely formidable without a precise particle localization technique. In this paper, we developed a fully-automated scheme to locate particles positions with high noise tolerance. Our scheme includes the following procedures: noise reduction using a hybrid Kalman filter method, particle segmentation based on a multiscale kernel graph cuts global and local segmentation algorithm, and a kinematic estimation based particle tracking method. Both isolated and partial-overlapped particles can be accurately identified with removal of unrelated pixels. Based on our quantitative analysis, 96.22% isolated particles and 84.19% partial-overlapped particles were successfully detected.
Zhu, Xiao Ran; Zhang, You Yun; Zhu, Yong Sheng [Xi' an Jiaotong Univ., Xi' an (China)
2012-09-15
Intelligent fault diagnosis benefits from efficient feature selection. Neighborhood rough sets are effective in feature selection. However, determining the neighborhood value accurately remains a challenge. The wrapper feature selection algorithm is designed by combining the kernel method and neighborhood rough sets to self-adaptively select sensitive features. The combination effectively solves the shortcomings in selecting the neighborhood value in the previous application process. The statistical features of time and frequency domains are used to describe the characteristic of the rolling bearing to make the intelligent fault diagnosis approach work. Three classification algorithms, namely, classification and regression tree (CART), commercial version 4.5 (C4.5), and radial basis function support vector machines (RBFSVM), are used to test UCI datasets and 10 fault datasets of rolling bearing. The results indicate that the diagnostic approach presented could effectively select the sensitive fault features and simultaneously identify the type and degree of the fault.
Zhu, Xiao Ran; Zhang, You Yun; Zhu, Yong Sheng
2012-01-01
Intelligent fault diagnosis benefits from efficient feature selection. Neighborhood rough sets are effective in feature selection. However, determining the neighborhood value accurately remains a challenge. The wrapper feature selection algorithm is designed by combining the kernel method and neighborhood rough sets to self-adaptively select sensitive features. The combination effectively solves the shortcomings in selecting the neighborhood value in the previous application process. The statistical features of time and frequency domains are used to describe the characteristic of the rolling bearing to make the intelligent fault diagnosis approach work. Three classification algorithms, namely, classification and regression tree (CART), commercial version 4.5 (C4.5), and radial basis function support vector machines (RBFSVM), are used to test UCI datasets and 10 fault datasets of rolling bearing. The results indicate that the diagnostic approach presented could effectively select the sensitive fault features and simultaneously identify the type and degree of the fault
Subspace confinement: how good is your qubit?
Devitt, Simon J; Schirmer, Sonia G; Oi, Daniel K L; Cole, Jared H; Hollenberg, Lloyd C L
2007-01-01
The basic operating element of standard quantum computation is the qubit, an isolated two-level system that can be accurately controlled, initialized and measured. However, the majority of proposed physical architectures for quantum computation are built from systems that contain much more complicated Hilbert space structures. Hence, defining a qubit requires the identification of an appropriate controllable two-dimensional sub-system. This prompts the obvious question of how well a qubit, thus defined, is confined to this subspace, and whether we can experimentally quantify the potential leakage into states outside the qubit subspace. We demonstrate how subspace leakage can be characterized using minimal theoretical assumptions by examining the Fourier spectrum of the oscillation experiment
Delong Feng
2016-05-01
Full Text Available Remaining useful life estimation of the prognostics and health management technique is a complicated and difficult research question for maintenance. In this article, we consider the problem of prognostics modeling and estimation of the turbofan engine under complicated circumstances and propose a kernel principal component analysis–based degradation model and remaining useful life estimation method for such aircraft engine. We first analyze the output data created by the turbofan engine thermodynamic simulation that is based on the kernel principal component analysis method and then distinguish the qualitative and quantitative relationships between the key factors. Next, we build a degradation model for the engine fault based on the following assumptions: the engine has only had constant failure (i.e. no sudden failure is included, and the engine has a Wiener process, which is a covariate stand for the engine system drift. To predict the remaining useful life of the turbofan engine, we built a health index based on the degradation model and used the method of maximum likelihood and the data from the thermodynamic simulation model to estimate the parameters of this degradation model. Through the data analysis, we obtained a trend model of the regression curve line that fits with the actual statistical data. Based on the predicted health index model and the data trend model, we estimate the remaining useful life of the aircraft engine as the index reaches zero. At last, a case study involving engine simulation data demonstrates the precision and performance advantages of this prediction method that we propose. At last, a case study involving engine simulation data demonstrates the precision and performance advantages of this proposed method, the precision of the method can reach to 98.9% and the average precision is 95.8%.
Monomial codes seen as invariant subspaces
García-Planas María Isabel
2017-08-01
Full Text Available It is well known that cyclic codes are very useful because of their applications, since they are not computationally expensive and encoding can be easily implemented. The relationship between cyclic codes and invariant subspaces is also well known. In this paper a generalization of this relationship is presented between monomial codes over a finite field and hyperinvariant subspaces of n under an appropriate linear transformation. Using techniques of Linear Algebra it is possible to deduce certain properties for this particular type of codes, generalizing known results on cyclic codes.
Matrix Krylov subspace methods for image restoration
khalide jbilou
2015-09-01
Full Text Available In the present paper, we consider some matrix Krylov subspace methods for solving ill-posed linear matrix equations and in those problems coming from the restoration of blurred and noisy images. Applying the well known Tikhonov regularization procedure leads to a Sylvester matrix equation depending the Tikhonov regularized parameter. We apply the matrix versions of the well known Krylov subspace methods, namely the Least Squared (LSQR and the conjugate gradient (CG methods to get approximate solutions representing the restored images. Some numerical tests are presented to show the effectiveness of the proposed methods.
Quantum Einstein gravity. Advancements of heat kernel-based renormalization group studies
Groh, Kai
2012-10-15
The asymptotic safety scenario allows to define a consistent theory of quantized gravity within the framework of quantum field theory. The central conjecture of this scenario is the existence of a non-Gaussian fixed point of the theory's renormalization group flow, that allows to formulate renormalization conditions that render the theory fully predictive. Investigations of this possibility use an exact functional renormalization group equation as a primary non-perturbative tool. This equation implements Wilsonian renormalization group transformations, and is demonstrated to represent a reformulation of the functional integral approach to quantum field theory. As its main result, this thesis develops an algebraic algorithm which allows to systematically construct the renormalization group flow of gauge theories as well as gravity in arbitrary expansion schemes. In particular, it uses off-diagonal heat kernel techniques to efficiently handle the non-minimal differential operators which appear due to gauge symmetries. The central virtue of the algorithm is that no additional simplifications need to be employed, opening the possibility for more systematic investigations of the emergence of non-perturbative phenomena. As a by-product several novel results on the heat kernel expansion of the Laplace operator acting on general gauge bundles are obtained. The constructed algorithm is used to re-derive the renormalization group flow of gravity in the Einstein-Hilbert truncation, showing the manifest background independence of the results. The well-studied Einstein-Hilbert case is further advanced by taking the effect of a running ghost field renormalization on the gravitational coupling constants into account. A detailed numerical analysis reveals a further stabilization of the found non-Gaussian fixed point. Finally, the proposed algorithm is applied to the case of higher derivative gravity including all curvature squared interactions. This establishes an improvement
Quantum Einstein gravity. Advancements of heat kernel-based renormalization group studies
Groh, Kai
2012-10-01
The asymptotic safety scenario allows to define a consistent theory of quantized gravity within the framework of quantum field theory. The central conjecture of this scenario is the existence of a non-Gaussian fixed point of the theory's renormalization group flow, that allows to formulate renormalization conditions that render the theory fully predictive. Investigations of this possibility use an exact functional renormalization group equation as a primary non-perturbative tool. This equation implements Wilsonian renormalization group transformations, and is demonstrated to represent a reformulation of the functional integral approach to quantum field theory. As its main result, this thesis develops an algebraic algorithm which allows to systematically construct the renormalization group flow of gauge theories as well as gravity in arbitrary expansion schemes. In particular, it uses off-diagonal heat kernel techniques to efficiently handle the non-minimal differential operators which appear due to gauge symmetries. The central virtue of the algorithm is that no additional simplifications need to be employed, opening the possibility for more systematic investigations of the emergence of non-perturbative phenomena. As a by-product several novel results on the heat kernel expansion of the Laplace operator acting on general gauge bundles are obtained. The constructed algorithm is used to re-derive the renormalization group flow of gravity in the Einstein-Hilbert truncation, showing the manifest background independence of the results. The well-studied Einstein-Hilbert case is further advanced by taking the effect of a running ghost field renormalization on the gravitational coupling constants into account. A detailed numerical analysis reveals a further stabilization of the found non-Gaussian fixed point. Finally, the proposed algorithm is applied to the case of higher derivative gravity including all curvature squared interactions. This establishes an improvement of
1986-03-01
A study on radiation dose control in packages of radioactive waste from nuclear facilities, hospitals and industries, such as sources of Ra-226, Co-60, Ir-192 and Cs-137, is presented. The MAPA and MAPAM computer codes, based on point Kernel theory for calculating doses of several source-shielding type configurations, aiming to assure the safe transport conditions for these sources, was developed. The validation of the code for point sources, using the values provided by NCRP, for the thickness of lead and concrete shieldings, limiting the dose at 100 Mrem/hr for several distances from the source to the detector, was carried out. The validation for non point sources was carried out, measuring experimentally radiation dose from packages developed by Brazilian CNEN/S.P. for removing the sources. (M.C.K.) [pt
Quantum Computing in Decoherence-Free Subspace Constructed by Triangulation
Bi, Qiao; Guo, Liu; Ruda, H. E.
2010-01-01
A formalism for quantum computing in decoherence-free subspaces is presented. The constructed subspaces are partial triangulated to an index related to environment. The quantum states in the subspaces are just projected states which are ruled by a subdynamic kinetic equation. These projected states can be used to perform ideal quantum logical operations without decoherence.
Quantum Computing in Decoherence-Free Subspace Constructed by Triangulation
Qiao Bi
2010-01-01
Full Text Available A formalism for quantum computing in decoherence-free subspaces is presented. The constructed subspaces are partial triangulated to an index related to environment. The quantum states in the subspaces are just projected states which are ruled by a subdynamic kinetic equation. These projected states can be used to perform ideal quantum logical operations without decoherence.
LBAS: Lanczos Bidiagonalization with Subspace Augmentation for Discrete Inverse Problems
Hansen, Per Christian; Abe, Kyniyoshi
The regularizing properties of Lanczos bidiagonalization are powerful when the underlying Krylov subspace captures the dominating components of the solution. In some applications the regularized solution can be further improved by augmenting the Krylov subspace with a low-dimensional subspace tha...
Rezaee, Malahat; Basri, Mahiran; Rahman, Raja Noor Zaliha Raja Abdul; Salleh, Abu Bakar; Chaibakhsh, Naz; Karjiban, Roghayeh Abedi
2014-01-01
Response surface methodology was employed to study the effect of formulation composition variables, water content (60%-80%, w/w) and oil and surfactant (O/S) ratio (0.17-1.33), as well as high-shear emulsification conditions, mixing rate (300-3,000 rpm) and mixing time (5-30 minutes) on the properties of sodium diclofenac-loaded palm kernel oil esters-nanoemulsions. The two response variables were droplet size and viscosity. Optimization of the conditions according to the four variables was performed for preparation of the nanoemulsions with the minimum values of particle size and viscosity. The results showed that the experimental data could be sufficiently fitted into a third-order polynomial model with multiple regression coefficients (R(2) ) of 0.938 and 0.994 for the particle size and viscosity, respectively. Water content, O/S ratio and mixing time, quadrics of all independent variables, interaction between O/S ratio and mixing rate and between mixing time and rate, as well as cubic term of water content had a significant effect (Pdiclofenac nanoemulsions were predicted to be: 71.36% water content; 0.69 O/S ratio; 950 rpm mixing rate, and 5 minute mixing time. The optimized formulation showed good storage stability in different temperatures.
Chen, Xing; Niu, Ya-Wei; Wang, Guang-Hui; Yan, Gui-Ying
2017-12-12
Recently, as the research of microRNA (miRNA) continues, there are plenty of experimental evidences indicating that miRNA could be associated with various human complex diseases development and progression. Hence, it is necessary and urgent to pay more attentions to the relevant study of predicting diseases associated miRNAs, which may be helpful for effective prevention, diagnosis and treatment of human diseases. Especially, constructing computational methods to predict potential miRNA-disease associations is worthy of more studies because of the feasibility and effectivity. In this work, we developed a novel computational model of multiple kernels learning-based Kronecker regularized least squares for MiRNA-disease association prediction (MKRMDA), which could reveal potential miRNA-disease associations by automatically optimizing the combination of multiple kernels for disease and miRNA. MKRMDA obtained AUCs of 0.9040 and 0.8446 in global and local leave-one-out cross validation, respectively. Meanwhile, MKRMDA achieved average AUCs of 0.8894 ± 0.0015 in fivefold cross validation. Furthermore, we conducted three different kinds of case studies on some important human cancers for further performance evaluation. In the case studies of colonic cancer, esophageal cancer and lymphoma based on known miRNA-disease associations in HMDDv2.0 database, 76, 94 and 88% of the corresponding top 50 predicted miRNAs were confirmed by experimental reports, respectively. In another two kinds of case studies for new diseases without any known associated miRNAs and diseases only with known associations in HMDDv1.0 database, the verified ratios of two different cancers were 88 and 94%, respectively. All the results mentioned above adequately showed the reliable prediction ability of MKRMDA. We anticipated that MKRMDA could serve to facilitate further developments in the field and the follow-up investigations by biomedical researchers.
Subspace System Identification of the Kalman Filter
David Di Ruscio
2003-07-01
Full Text Available Some proofs concerning a subspace identification algorithm are presented. It is proved that the Kalman filter gain and the noise innovations process can be identified directly from known input and output data without explicitly solving the Riccati equation. Furthermore, it is in general and for colored inputs, proved that the subspace identification of the states only is possible if the deterministic part of the system is known or identified beforehand. However, if the inputs are white, then, it is proved that the states can be identified directly. Some alternative projection matrices which can be used to compute the extended observability matrix directly from the data are presented. Furthermore, an efficient method for computing the deterministic part of the system is presented. The closed loop subspace identification problem is also addressed and it is shown that this problem is solved and unbiased estimates are obtained by simply including a filter in the feedback. Furthermore, an algorithm for consistent closed loop subspace estimation is presented. This algorithm is using the controller parameters in order to overcome the bias problem.
Quantum Zeno subspaces induced by temperature
Militello, B.; Scala, M.; Messina, A. [Dipartimento di Fisica dell' Universita di Palermo, Via Archirafi 36, I-90123 Palermo (Italy)
2011-08-15
We discuss the partitioning of the Hilbert space of a quantum system induced by the interaction with another system at thermal equilibrium, showing that the higher the temperature the more effective is the formation of Zeno subspaces. We show that our analysis keeps its validity even in the case of interaction with a bosonic reservoir, provided appropriate limitations of the relevant bandwidth.
Counting Subspaces of a Finite Vector Space
Home; Journals; Resonance – Journal of Science Education; Volume 15; Issue 11. Counting Subspaces of a Finite Vector Space – 1. Amritanshu Prasad. General Article Volume 15 Issue 11 November 2010 pp 977-987. Fulltext. Click here to view fulltext PDF. Permanent link:
RTOS kernel in portable electrocardiograph
Centeno, C. A.; Voos, J. A.; Riva, G. G.; Zerbini, C.; Gonzalez, E. A.
2011-12-01
This paper presents the use of a Real Time Operating System (RTOS) on a portable electrocardiograph based on a microcontroller platform. All medical device digital functions are performed by the microcontroller. The electrocardiograph CPU is based on the 18F4550 microcontroller, in which an uCOS-II RTOS can be embedded. The decision associated with the kernel use is based on its benefits, the license for educational use and its intrinsic time control and peripherals management. The feasibility of its use on the electrocardiograph is evaluated based on the minimum memory requirements due to the kernel structure. The kernel's own tools were used for time estimation and evaluation of resources used by each process. After this feasibility analysis, the migration from cyclic code to a structure based on separate processes or tasks able to synchronize events is used; resulting in an electrocardiograph running on one Central Processing Unit (CPU) based on RTOS.
RTOS kernel in portable electrocardiograph
Centeno, C A; Voos, J A; Riva, G G; Zerbini, C; Gonzalez, E A
2011-01-01
This paper presents the use of a Real Time Operating System (RTOS) on a portable electrocardiograph based on a microcontroller platform. All medical device digital functions are performed by the microcontroller. The electrocardiograph CPU is based on the 18F4550 microcontroller, in which an uCOS-II RTOS can be embedded. The decision associated with the kernel use is based on its benefits, the license for educational use and its intrinsic time control and peripherals management. The feasibility of its use on the electrocardiograph is evaluated based on the minimum memory requirements due to the kernel structure. The kernel's own tools were used for time estimation and evaluation of resources used by each process. After this feasibility analysis, the migration from cyclic code to a structure based on separate processes or tasks able to synchronize events is used; resulting in an electrocardiograph running on one Central Processing Unit (CPU) based on RTOS.
Rashidi, Nor Adilla; Yusup, Suzana
2018-05-09
The feasibility of biomass-based activated carbons has received a huge attention due to their excellent characteristics such as inexpensiveness, good adsorption behaviour and potential to reduce a strong dependency towards non-renewable precursors. Therefore, in this research work, eco-friendly activated carbon from palm kernel shell that has been produced from one-stage physical activation by using the Box-Behnken design of Response Surface Methodology is highlighted. The effect of three input parameters-temperature, dwell time and gas flow rate-towards product yield and carbon dioxide (CO 2 ) uptake at room temperature and atmospheric pressure are studied. Model accuracy has been evaluated through the ANOVA analysis and lack-of-fit test. Accordingly, the optimum condition in synthesising the activated carbon with adequate CO 2 adsorption capacity of 2.13 mmol/g and product yield of 25.15 wt% is found at a temperature of 850 °C, holding time of 60 min and CO 2 flow rate of 450 cm 3 /min. The synthesised activated carbon has been characterised by diverse analytical instruments including thermogravimetric analyser, scanning electron microscope, as well as N 2 adsorption-desorption isotherm. The characterisation analysis indicates that the synthesised activated carbon has higher textural characteristics and porosity, together with better thermal stability and carbon content as compared to pristine palm kernel shell. Activated carbon production via one-step activation approach is economical since its carbon yield is within the industrial target, whereas CO 2 uptake is comparable to the synthesised activated carbon from conventional dual-stage activation, commercial activated carbon and other published data from literature.
Surface EMG decomposition based on K-means clustering and convolution kernel compensation.
Ning, Yong; Zhu, Xiangjun; Zhu, Shanan; Zhang, Yingchun
2015-03-01
A new approach has been developed by combining the K-mean clustering (KMC) method and a modified convolution kernel compensation (CKC) method for multichannel surface electromyogram (EMG) decomposition. The KMC method was first utilized to cluster vectors of observations at different time instants and then estimate the initial innervation pulse train (IPT). The CKC method, modified with a novel multistep iterative process, was conducted to update the estimated IPT. The performance of the proposed K-means clustering-Modified CKC (KmCKC) approach was evaluated by reconstructing IPTs from both simulated and experimental surface EMG signals. The KmCKC approach successfully reconstructed all 10 IPTs from the simulated surface EMG signals with true positive rates (TPR) of over 90% with a low signal-to-noise ratio (SNR) of -10 dB. More than 10 motor units were also successfully extracted from the 64-channel experimental surface EMG signals of the first dorsal interosseous (FDI) muscles when a contraction force was held at 8 N by using the KmCKC approach. A "two-source" test was further conducted with 64-channel surface EMG signals. The high percentage of common MUs and common pulses (over 92% at all force levels) between the IPTs reconstructed from the two independent groups of surface EMG signals demonstrates the reliability and capability of the proposed KmCKC approach in multichannel surface EMG decomposition. Results from both simulated and experimental data are consistent and confirm that the proposed KmCKC approach can successfully reconstruct IPTs with high accuracy at different levels of contraction.
Wang, Yongcui; Chen, Shilong; Deng, Naiyang; Wang, Yong
2013-01-01
Computational inference of novel therapeutic values for existing drugs, i.e., drug repositioning, offers the great prospect for faster and low-risk drug development. Previous researches have indicated that chemical structures, target proteins, and side-effects could provide rich information in drug similarity assessment and further disease similarity. However, each single data source is important in its own way and data integration holds the great promise to reposition drug more accurately. Here, we propose a new method for drug repositioning, PreDR (Predict Drug Repositioning), to integrate molecular structure, molecular activity, and phenotype data. Specifically, we characterize drug by profiling in chemical structure, target protein, and side-effects space, and define a kernel function to correlate drugs with diseases. Then we train a support vector machine (SVM) to computationally predict novel drug-disease interactions. PreDR is validated on a well-established drug-disease network with 1,933 interactions among 593 drugs and 313 diseases. By cross-validation, we find that chemical structure, drug target, and side-effects information are all predictive for drug-disease relationships. More experimentally observed drug-disease interactions can be revealed by integrating these three data sources. Comparison with existing methods demonstrates that PreDR is competitive both in accuracy and coverage. Follow-up database search and pathway analysis indicate that our new predictions are worthy of further experimental validation. Particularly several novel predictions are supported by clinical trials databases and this shows the significant prospects of PreDR in future drug treatment. In conclusion, our new method, PreDR, can serve as a useful tool in drug discovery to efficiently identify novel drug-disease interactions. In addition, our heterogeneous data integration framework can be applied to other problems. PMID:24244318
Conjunctive patches subspace learning with side information for collaborative image retrieval.
Zhang, Lining; Wang, Lipo; Lin, Weisi
2012-08-01
Content-Based Image Retrieval (CBIR) has attracted substantial attention during the past few years for its potential practical applications to image management. A variety of Relevance Feedback (RF) schemes have been designed to bridge the semantic gap between the low-level visual features and the high-level semantic concepts for an image retrieval task. Various Collaborative Image Retrieval (CIR) schemes aim to utilize the user historical feedback log data with similar and dissimilar pairwise constraints to improve the performance of a CBIR system. However, existing subspace learning approaches with explicit label information cannot be applied for a CIR task, although the subspace learning techniques play a key role in various computer vision tasks, e.g., face recognition and image classification. In this paper, we propose a novel subspace learning framework, i.e., Conjunctive Patches Subspace Learning (CPSL) with side information, for learning an effective semantic subspace by exploiting the user historical feedback log data for a CIR task. The CPSL can effectively integrate the discriminative information of labeled log images, the geometrical information of labeled log images and the weakly similar information of unlabeled images together to learn a reliable subspace. We formally formulate this problem into a constrained optimization problem and then present a new subspace learning technique to exploit the user historical feedback log data. Extensive experiments on both synthetic data sets and a real-world image database demonstrate the effectiveness of the proposed scheme in improving the performance of a CBIR system by exploiting the user historical feedback log data.
Pure endmember extraction using robust kernel archetypoid analysis for hyperspectral imagery
Sun, Weiwei; Yang, Gang; Wu, Ke; Li, Weiyue; Zhang, Dianfa
2017-09-01
A robust kernel archetypoid analysis (RKADA) method is proposed to extract pure endmembers from hyperspectral imagery (HSI). The RKADA assumes that each pixel is a sparse linear mixture of all endmembers and each endmember corresponds to a real pixel in the image scene. First, it improves the re8gular archetypal analysis with a new binary sparse constraint, and the adoption of the kernel function constructs the principal convex hull in an infinite Hilbert space and enlarges the divergences between pairwise pixels. Second, the RKADA transfers the pure endmember extraction problem into an optimization problem by minimizing residual errors with the Huber loss function. The Huber loss function reduces the effects from big noises and outliers in the convergence procedure of RKADA and enhances the robustness of the optimization function. Third, the random kernel sinks for fast kernel matrix approximation and the two-stage algorithm for optimizing initial pure endmembers are utilized to improve its computational efficiency in realistic implementations of RKADA, respectively. The optimization equation of RKADA is solved by using the block coordinate descend scheme and the desired pure endmembers are finally obtained. Six state-of-the-art pure endmember extraction methods are employed to make comparisons with the RKADA on both synthetic and real Cuprite HSI datasets, including three geometrical algorithms vertex component analysis (VCA), alternative volume maximization (AVMAX) and orthogonal subspace projection (OSP), and three matrix factorization algorithms the preconditioning for successive projection algorithm (PreSPA), hierarchical clustering based on rank-two nonnegative matrix factorization (H2NMF) and self-dictionary multiple measurement vector (SDMMV). Experimental results show that the RKADA outperforms all the six methods in terms of spectral angle distance (SAD) and root-mean-square-error (RMSE). Moreover, the RKADA has short computational times in offline
Chen, Zhicong; Wu, Lijun; Cheng, Shuying; Lin, Peijie; Wu, Yue; Lin, Wencheng
2017-01-01
Highlights: •An improved Simulink based modeling method is proposed for PV modules and arrays. •Key points of I-V curves and PV model parameters are used as the feature variables. •Kernel extreme learning machine (KELM) is explored for PV arrays fault diagnosis. •The parameters of KELM algorithm are optimized by the Nelder-Mead simplex method. •The optimized KELM fault diagnosis model achieves high accuracy and reliability. -- Abstract: Fault diagnosis of photovoltaic (PV) arrays is important for improving the reliability, efficiency and safety of PV power stations, because the PV arrays usually operate in harsh outdoor environment and tend to suffer various faults. Due to the nonlinear output characteristics and varying operating environment of PV arrays, many machine learning based fault diagnosis methods have been proposed. However, there still exist some issues: fault diagnosis performance is still limited due to insufficient monitored information; fault diagnosis models are not efficient to be trained and updated; labeled fault data samples are hard to obtain by field experiments. To address these issues, this paper makes contribution in the following three aspects: (1) based on the key points and model parameters extracted from monitored I-V characteristic curves and environment condition, an effective and efficient feature vector of seven dimensions is proposed as the input of the fault diagnosis model; (2) the emerging kernel based extreme learning machine (KELM), which features extremely fast learning speed and good generalization performance, is utilized to automatically establish the fault diagnosis model. Moreover, the Nelder-Mead Simplex (NMS) optimization method is employed to optimize the KELM parameters which affect the classification performance; (3) an improved accurate Simulink based PV modeling approach is proposed for a laboratory PV array to facilitate the fault simulation and data sample acquisition. Intensive fault experiments are
Extending the subspace hybrid method for eigenvalue problems in reactor physics calculation
Zhang, Q.; Abdel-Khalik, H. S.
2013-01-01
This paper presents an innovative hybrid Monte-Carlo-Deterministic method denoted by the SUBSPACE method designed for improving the efficiency of hybrid methods for reactor analysis applications. The SUBSPACE method achieves its high computational efficiency by taking advantage of the existing correlations between desired responses. Recently, significant gains in computational efficiency have been demonstrated using this method for source driven problems. Within this work the mathematical theory behind the SUBSPACE method is introduced and extended to address core wide level k-eigenvalue problems. The method's efficiency is demonstrated based on a three-dimensional quarter-core problem, where responses are sought on the pin cell level. The SUBSPACE method is compared to the FW-CADIS method and is found to be more efficient for the utilized test problem because of the reason that the FW-CADIS method solves a forward eigenvalue problem and an adjoint fixed-source problem while the SUBSPACE method only solves an adjoint fixed-source problem. Based on the favorable results obtained here, we are confident that the applicability of Monte Carlo for large scale reactor analysis could be realized in the near future. (authors)
Kernel Principal Component Analysis for dimensionality reduction in fMRI-based diagnosis of ADHD
Gagan S Sidhu
2012-11-01
Full Text Available This article explores various preprocessing tools that select/create features to help a learner produce a classifier that can use fMRI data to effectively discriminate Attention-Deficit Hyperactivity Disorder (ADHD patients from healthy controls. We consider four different learning tasks: predicting either two (ADHD vs control or three classes (ADHD-1 vs ADHD-3 vs control, where each use either the imaging data only, or the phenotypic and imaging data. After averaging, BOLD-signal normalization, and masking of the fMRI images, we considered applying Fast Fourier Transform (FFT, possibly followed by some Principal Component Analysis (PCA variant (over time: PCA-t; over space and time: PCA-st or the kernelized variant, kPCA-st, to produce inputs to a learner, to determine which learned classifier performs the best – or at least better than the baseline of 64.2%, which is the proportion of the majority class (here, controls.In the two-class setting, PCA-t and PCA-st did not perform statistically better than baseline, whereas FFT and kPCA-st did (FFT, 68.4%; kPCA-st, 70.3%; when combined with the phenotypic data, which by itself produces 72.9% accuracy, all methods performed statistically better than the baseline, but none did better than using the phenotypic data. In the three-class setting, neither the PCA variants, or the phenotypic data classifiers, performed statistically better than the baseline.We next used the FFT output as input to the PCA variants. In the two-class setting, the PCA variants performed statistically better than the baseline using either the FFTed waveforms only (FFT+PCA-t, 69.6%,; FFT+PCA-st, 69.3% ; FFT+kPCA-st, 68.7%, or combining them with the phenotypic data (FFT+PCA-t, 70.6%; FFT+PCA-st, 70.6%; kPCA-st, 76%. In both settings, combining FFT+kPCA-st’s features with the phenotypic data was better than using only the phenotypic data, with the result in the two-class setting being statistically better.
Partial Deconvolution with Inaccurate Blur Kernel.
Ren, Dongwei; Zuo, Wangmeng; Zhang, David; Xu, Jun; Zhang, Lei
2017-10-17
Most non-blind deconvolution methods are developed under the error-free kernel assumption, and are not robust to inaccurate blur kernel. Unfortunately, despite the great progress in blind deconvolution, estimation error remains inevitable during blur kernel estimation. Consequently, severe artifacts such as ringing effects and distortions are likely to be introduced in the non-blind deconvolution stage. In this paper, we tackle this issue by suggesting: (i) a partial map in the Fourier domain for modeling kernel estimation error, and (ii) a partial deconvolution model for robust deblurring with inaccurate blur kernel. The partial map is constructed by detecting the reliable Fourier entries of estimated blur kernel. And partial deconvolution is applied to wavelet-based and learning-based models to suppress the adverse effect of kernel estimation error. Furthermore, an E-M algorithm is developed for estimating the partial map and recovering the latent sharp image alternatively. Experimental results show that our partial deconvolution model is effective in relieving artifacts caused by inaccurate blur kernel, and can achieve favorable deblurring quality on synthetic and real blurry images.Most non-blind deconvolution methods are developed under the error-free kernel assumption, and are not robust to inaccurate blur kernel. Unfortunately, despite the great progress in blind deconvolution, estimation error remains inevitable during blur kernel estimation. Consequently, severe artifacts such as ringing effects and distortions are likely to be introduced in the non-blind deconvolution stage. In this paper, we tackle this issue by suggesting: (i) a partial map in the Fourier domain for modeling kernel estimation error, and (ii) a partial deconvolution model for robust deblurring with inaccurate blur kernel. The partial map is constructed by detecting the reliable Fourier entries of estimated blur kernel. And partial deconvolution is applied to wavelet-based and learning-based
Yu-xin Zhao
2014-01-01
Full Text Available This paper presents a novel wavelet kernel neural network (WKNN with wavelet kernel function. It is applicable in online learning with adaptive parameters and is applied on parameters tuning of fractional-order PID (FOPID controller, which could handle time delay problem of the complex control system. Combining the wavelet function and the kernel function, the wavelet kernel function is adopted and validated the availability for neural network. Compared to the conservative wavelet neural network, the most innovative character of the WKNN is its rapid convergence and high precision in parameters updating process. Furthermore, the integrated pressurized water reactor (IPWR system is established by RELAP5, and a novel control strategy combining WKNN and fuzzy logic rule is proposed for shortening controlling time and utilizing the experiential knowledge sufficiently. Finally, experiment results verify that the control strategy and controller proposed have the practicability and reliability in actual complicated system.
Jinlu Sheng
2016-07-01
Full Text Available To effectively extract the typical features of the bearing, a new method that related the local mean decomposition Shannon entropy and improved kernel principal component analysis model was proposed. First, the features are extracted by time–frequency domain method, local mean decomposition, and using the Shannon entropy to process the original separated product functions, so as to get the original features. However, the features been extracted still contain superfluous information; the nonlinear multi-features process technique, kernel principal component analysis, is introduced to fuse the characters. The kernel principal component analysis is improved by the weight factor. The extracted characteristic features were inputted in the Morlet wavelet kernel support vector machine to get the bearing running state classification model, bearing running state was thereby identified. Cases of test and actual were analyzed.
View subspaces for indexing and retrieval of 3D models
Dutagaci, Helin; Godil, Afzal; Sankur, Bülent; Yemez, Yücel
2010-02-01
View-based indexing schemes for 3D object retrieval are gaining popularity since they provide good retrieval results. These schemes are coherent with the theory that humans recognize objects based on their 2D appearances. The viewbased techniques also allow users to search with various queries such as binary images, range images and even 2D sketches. The previous view-based techniques use classical 2D shape descriptors such as Fourier invariants, Zernike moments, Scale Invariant Feature Transform-based local features and 2D Digital Fourier Transform coefficients. These methods describe each object independent of others. In this work, we explore data driven subspace models, such as Principal Component Analysis, Independent Component Analysis and Nonnegative Matrix Factorization to describe the shape information of the views. We treat the depth images obtained from various points of the view sphere as 2D intensity images and train a subspace to extract the inherent structure of the views within a database. We also show the benefit of categorizing shapes according to their eigenvalue spread. Both the shape categorization and data-driven feature set conjectures are tested on the PSB database and compared with the competitor view-based 3D shape retrieval algorithms.
Clustering via Kernel Decomposition
Have, Anna Szynkowiak; Girolami, Mark A.; Larsen, Jan
2006-01-01
Methods for spectral clustering have been proposed recently which rely on the eigenvalue decomposition of an affinity matrix. In this work it is proposed that the affinity matrix is created based on the elements of a non-parametric density estimator. This matrix is then decomposed to obtain...... posterior probabilities of class membership using an appropriate form of nonnegative matrix factorization. The troublesome selection of hyperparameters such as kernel width and number of clusters can be obtained using standard cross-validation methods as is demonstrated on a number of diverse data sets....
SU-F-T-672: A Novel Kernel-Based Dose Engine for KeV Photon Beams
Reinhart, M; Fast, M F; Nill, S; Oelfke, U [The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London (United Kingdom)
2016-06-15
Purpose: Mimicking state-of-the-art patient radiotherapy with high precision irradiators for small animals allows advanced dose-effect studies and radiobiological investigations. One example is the implementation of pre-clinical IMRT-like irradiations, which requires the development of inverse planning for keV photon beams. As a first step, we present a novel kernel-based dose calculation engine for keV x-rays with explicit consideration of energy and material dependencies. Methods: We follow a superposition-convolution approach adapted to keV x-rays, based on previously published work on micro-beam therapy. In small animal radiotherapy, we assume local energy deposition at the photon interaction point, since the electron ranges in tissue are of the same order of magnitude as the voxel size. This allows us to use photon-only kernel sets generated by MC simulations, which are pre-calculated for six energy windows and ten base materials. We validate our stand-alone dose engine against Geant4 MC simulations for various beam configurations in water, slab phantoms with bone and lung inserts, and on a mouse CT with (0.275mm)3 voxels. Results: We observe good agreement for all cases. For field sizes of 1mm{sup 2} to 1cm{sup 2} in water, the depth dose curves agree within 1% (mean), with the largest deviations in the first voxel (4%) and at depths>5cm (<2.5%). The out-of-field doses at 1cm depth agree within 8% (mean) for all but the smallest field size. In slab geometries, the mean agreement was within 3%, with maximum deviations of 8% at water-bone interfaces. The γ-index (1mm/1%) passing rate for a single-field mouse irradiation is 71%. Conclusion: The presented dose engine yields an accurate representation of keV-photon doses suitable for inverse treatment planning for IMRT. It has the potential to become a significantly faster yet sufficiently accurate alternative to full MC simulations. Further investigations will focus on energy sampling as well as calculation
Alam, Md. Ashad; Fukumizu, Kenji; Wang, Yu-Ping
2016-01-01
To the best of our knowledge, there are no general well-founded robust methods for statistical unsupervised learning. Most of the unsupervised methods explicitly or implicitly depend on the kernel covariance operator (kernel CO) or kernel cross-covariance operator (kernel CCO). They are sensitive to contaminated data, even when using bounded positive definite kernels. First, we propose robust kernel covariance operator (robust kernel CO) and robust kernel crosscovariance operator (robust kern...
Hekmatmanesh, Amin; Jamaloo, Fatemeh; Wu, Huapeng; Handroos, Heikki; Kilpeläinen, Asko
2018-04-01
Brain Computer Interface (BCI) can be a challenge for developing of robotic, prosthesis and human-controlled systems. This work focuses on the implementation of a common spatial pattern (CSP) base algorithm to detect event related desynchronization patterns. Utilizing famous previous work in this area, features are extracted by filter bank with common spatial pattern (FBCSP) method, and then weighted by a sensitive learning vector quantization (SLVQ) algorithm. In the current work, application of the radial basis function (RBF) as a mapping kernel of linear discriminant analysis (KLDA) method on the weighted features, allows the transfer of data into a higher dimension for more discriminated data scattering by RBF kernel. Afterwards, support vector machine (SVM) with generalized radial basis function (GRBF) kernel is employed to improve the efficiency and robustness of the classification. Averagely, 89.60% accuracy and 74.19% robustness are achieved. BCI Competition III, Iva data set is used to evaluate the algorithm for detecting right hand and foot imagery movement patterns. Results show that combination of KLDA with SVM-GRBF classifier makes 8.9% and 14.19% improvements in accuracy and robustness, respectively. For all the subjects, it is concluded that mapping the CSP features into a higher dimension by RBF and utilization GRBF as a kernel of SVM, improve the accuracy and reliability of the proposed method.
A subspace approach to high-resolution spectroscopic imaging.
Lam, Fan; Liang, Zhi-Pei
2014-04-01
To accelerate spectroscopic imaging using sparse sampling of (k,t)-space and subspace (or low-rank) modeling to enable high-resolution metabolic imaging with good signal-to-noise ratio. The proposed method, called SPectroscopic Imaging by exploiting spatiospectral CorrElation, exploits a unique property known as partial separability of spectroscopic signals. This property indicates that high-dimensional spectroscopic signals reside in a very low-dimensional subspace and enables special data acquisition and image reconstruction strategies to be used to obtain high-resolution spatiospectral distributions with good signal-to-noise ratio. More specifically, a hybrid chemical shift imaging/echo-planar spectroscopic imaging pulse sequence is proposed for sparse sampling of (k,t)-space, and a low-rank model-based algorithm is proposed for subspace estimation and image reconstruction from sparse data with the capability to incorporate prior information and field inhomogeneity correction. The performance of the proposed method has been evaluated using both computer simulations and phantom studies, which produced very encouraging results. For two-dimensional spectroscopic imaging experiments on a metabolite phantom, a factor of 10 acceleration was achieved with a minimal loss in signal-to-noise ratio compared to the long chemical shift imaging experiments and with a significant gain in signal-to-noise ratio compared to the accelerated echo-planar spectroscopic imaging experiments. The proposed method, SPectroscopic Imaging by exploiting spatiospectral CorrElation, is able to significantly accelerate spectroscopic imaging experiments, making high-resolution metabolic imaging possible. Copyright © 2014 Wiley Periodicals, Inc.
Panel data specifications in nonparametric kernel regression
Czekaj, Tomasz Gerard; Henningsen, Arne
parametric panel data estimators to analyse the production technology of Polish crop farms. The results of our nonparametric kernel regressions generally differ from the estimates of the parametric models but they only slightly depend on the choice of the kernel functions. Based on economic reasoning, we...
Improving the Bandwidth Selection in Kernel Equating
Andersson, Björn; von Davier, Alina A.
2014-01-01
We investigate the current bandwidth selection methods in kernel equating and propose a method based on Silverman's rule of thumb for selecting the bandwidth parameters. In kernel equating, the bandwidth parameters have previously been obtained by minimizing a penalty function. This minimization process has been criticized by practitioners…
Shi, Jun; Liu, Xiao; Li, Yan; Zhang, Qi; Li, Yingjie; Ying, Shihui
2015-10-30
Electroencephalography (EEG) based sleep staging is commonly used in clinical routine. Feature extraction and representation plays a crucial role in EEG-based automatic classification of sleep stages. Sparse representation (SR) is a state-of-the-art unsupervised feature learning method suitable for EEG feature representation. Collaborative representation (CR) is an effective data coding method used as a classifier. Here we use CR as a data representation method to learn features from the EEG signal. A joint collaboration model is established to develop a multi-view learning algorithm, and generate joint CR (JCR) codes to fuse and represent multi-channel EEG signals. A two-stage multi-view learning-based sleep staging framework is then constructed, in which JCR and joint sparse representation (JSR) algorithms first fuse and learning the feature representation from multi-channel EEG signals, respectively. Multi-view JCR and JSR features are then integrated and sleep stages recognized by a multiple kernel extreme learning machine (MK-ELM) algorithm with grid search. The proposed two-stage multi-view learning algorithm achieves superior performance for sleep staging. With a K-means clustering based dictionary, the mean classification accuracy, sensitivity and specificity are 81.10 ± 0.15%, 71.42 ± 0.66% and 94.57 ± 0.07%, respectively; while with the dictionary learned using the submodular optimization method, they are 80.29 ± 0.22%, 71.26 ± 0.78% and 94.38 ± 0.10%, respectively. The two-stage multi-view learning based sleep staging framework outperforms all other classification methods compared in this work, while JCR is superior to JSR. The proposed multi-view learning framework has the potential for sleep staging based on multi-channel or multi-modality polysomnography signals. Copyright © 2015 Elsevier B.V. All rights reserved.
Matsumoto, S.; Ohno, Y.; Takenaka, D.; Sugimura, K.; Yamagata, H.
2007-01-01
Classification of the nodule candidates in computer-aided detection (CAD) of lung nodules in CT images was addressed by constructing a nonlinear discriminant function using a kernel-based learning algorithm called the kernel recursive least-squares (KRLS) algorithm. Using the nodule candidates derived from the processing by a CAD scheme of 100 CT datasets containing 253 non-calcified nodules or 3 mm or larger as determined by the consensus of two thoracic radiologists, the following trial were carried out 100 times: by randomly selecting 50 datasets for training, a nonlinear discriminant function was obtained using the nodule candidates in the training datasets and tested with the remaining candidates; for comparison, a rule-based classification was tested in a similar manner. At the number of false positives per case of about 5, the nonlinear classification method showed an improved sensitivity of 80% (mean over the 100 trials) compared with 74% of the rule-based method. (orig.)
Subspace methods for identification of human ankle joint stiffness.
Zhao, Y; Westwick, D T; Kearney, R E
2011-11-01
Joint stiffness, the dynamic relationship between the angular position of a joint and the torque acting about it, describes the dynamic, mechanical behavior of a joint during posture and movement. Joint stiffness arises from both intrinsic and reflex mechanisms, but the torques due to these mechanisms cannot be measured separately experimentally, since they appear and change together. Therefore, the direct estimation of the intrinsic and reflex stiffnesses is difficult. In this paper, we present a new, two-step procedure to estimate the intrinsic and reflex components of ankle stiffness. In the first step, a discrete-time, subspace-based method is used to estimate a state-space model for overall stiffness from the measured overall torque and then predict the intrinsic and reflex torques. In the second step, continuous-time models for the intrinsic and reflex stiffnesses are estimated from the predicted intrinsic and reflex torques. Simulations and experimental results demonstrate that the algorithm estimates the intrinsic and reflex stiffnesses accurately. The new subspace-based algorithm has three advantages over previous algorithms: 1) It does not require iteration, and therefore, will always converge to an optimal solution; 2) it provides better estimates for data with high noise or short sample lengths; and 3) it provides much more accurate results for data acquired under the closed-loop conditions, that prevail when subjects interact with compliant loads.
Enhancing Low-Rank Subspace Clustering by Manifold Regularization.
Liu, Junmin; Chen, Yijun; Zhang, JiangShe; Xu, Zongben
2014-07-25
Recently, low-rank representation (LRR) method has achieved great success in subspace clustering (SC), which aims to cluster the data points that lie in a union of low-dimensional subspace. Given a set of data points, LRR seeks the lowest rank representation among the many possible linear combinations of the bases in a given dictionary or in terms of the data itself. However, LRR only considers the global Euclidean structure, while the local manifold structure, which is often important for many real applications, is ignored. In this paper, to exploit the local manifold structure of the data, a manifold regularization characterized by a Laplacian graph has been incorporated into LRR, leading to our proposed Laplacian regularized LRR (LapLRR). An efficient optimization procedure, which is based on alternating direction method of multipliers (ADMM), is developed for LapLRR. Experimental results on synthetic and real data sets are presented to demonstrate that the performance of LRR has been enhanced by using the manifold regularization.
Qiang Shang
Full Text Available Short-term traffic flow prediction is one of the most important issues in the field of intelligent transport system (ITS. Because of the uncertainty and nonlinearity, short-term traffic flow prediction is a challenging task. In order to improve the accuracy of short-time traffic flow prediction, a hybrid model (SSA-KELM is proposed based on singular spectrum analysis (SSA and kernel extreme learning machine (KELM. SSA is used to filter out the noise of traffic flow time series. Then, the filtered traffic flow data is used to train KELM model, the optimal input form of the proposed model is determined by phase space reconstruction, and parameters of the model are optimized by gravitational search algorithm (GSA. Finally, case validation is carried out using the measured data of an expressway in Xiamen, China. And the SSA-KELM model is compared with several well-known prediction models, including support vector machine, extreme learning machine, and single KLEM model. The experimental results demonstrate that performance of the proposed model is superior to that of the comparison models. Apart from accuracy improvement, the proposed model is more robust.
Asger Nyman Christiansen
2012-11-01
Full Text Available Currently, no objective method exists for estimating the rate of change in the colour of meat. Consequently, the purpose of this work is to develop a procedure capable of monitoring the change in colour of meat over time, environment and ingredients. This provides a useful tool to determine which storage environments and ingredients a manufacturer should add to meat to reduce the rate of change in colour. The procedure consists of taking multi-spectral images of a piece of meat as a function of time, clustering the pixels of these images into categories, including several types of meat, and extracting colour information from each category. The focus has primarily been on achieving an accurate categorisation since this is crucial to develop a useful method. The categorisation is done by applying an orthogonal transformation followed by k-means clustering. The purpose of the orthogonal transformation is to reduce the noise and amount of data while enhancing the difference between the categories. The orthogonal transformations principal components analysis, minimum noise fraction analysis and kernel-based versions of these have been applied to test which produce the most accurate categorisation.
Cho, Dongrae; Ham, Jinsil; Oh, Jooyoung; Park, Jeanho; Kim, Sayup; Lee, Nak-Kyu; Lee, Boreom
2017-10-24
Virtual reality (VR) is a computer technique that creates an artificial environment composed of realistic images, sounds, and other sensations. Many researchers have used VR devices to generate various stimuli, and have utilized them to perform experiments or to provide treatment. In this study, the participants performed mental tasks using a VR device while physiological signals were measured: a photoplethysmogram (PPG), electrodermal activity (EDA), and skin temperature (SKT). In general, stress is an important factor that can influence the autonomic nervous system (ANS). Heart-rate variability (HRV) is known to be related to ANS activity, so we used an HRV derived from the PPG peak interval. In addition, the peak characteristics of the skin conductance (SC) from EDA and SKT variation can also reflect ANS activity; we utilized them as well. Then, we applied a kernel-based extreme-learning machine (K-ELM) to correctly classify the stress levels induced by the VR task to reflect five different levels of stress situations: baseline, mild stress, moderate stress, severe stress, and recovery. Twelve healthy subjects voluntarily participated in the study. Three physiological signals were measured in stress environment generated by VR device. As a result, the average classification accuracy was over 95% using K-ELM and the integrated feature (IT = HRV + SC + SKT). In addition, the proposed algorithm can embed a microcontroller chip since K-ELM algorithm have very short computation time. Therefore, a compact wearable device classifying stress levels using physiological signals can be developed.
Yao, Yuchen; Bao, Jie; Skyllas-Kazacos, Maria; Welch, Barry J.; Akhmetov, Sergey
2018-04-01
Individual anode current signals in aluminum reduction cells provide localized cell conditions in the vicinity of each anode, which contain more information than the conventionally measured cell voltage and line current. One common use of this measurement is to identify process faults that can cause significant changes in the anode current signals. While this method is simple and direct, it ignores the interactions between anode currents and other important process variables. This paper presents an approach that applies multivariate statistical analysis techniques to individual anode currents and other process operating data, for the detection and diagnosis of local process abnormalities in aluminum reduction cells. Specifically, since the Hall-Héroult process is time-varying with its process variables dynamically and nonlinearly correlated, dynamic kernel principal component analysis with moving windows is used. The cell is discretized into a number of subsystems, with each subsystem representing one anode and cell conditions in its vicinity. The fault associated with each subsystem is identified based on multivariate statistical control charts. The results show that the proposed approach is able to not only effectively pinpoint the problematic areas in the cell, but also assess the effect of the fault on different parts of the cell.
Galindo, I.; Romero, M. C.; Sánchez, N.; Morales, J. M.
2016-06-01
Risk management stakeholders in high-populated volcanic islands should be provided with the latest high-quality volcanic information. We present here the first volcanic susceptibility map of Lanzarote and Chinijo Islands and their submarine flanks based on updated chronostratigraphical and volcano structural data, as well as on the geomorphological analysis of the bathymetric data of the submarine flanks. The role of the structural elements in the volcanic susceptibility analysis has been reviewed: vents have been considered since they indicate where previous eruptions took place; eruptive fissures provide information about the stress field as they are the superficial expression of the dyke conduit; eroded dykes have been discarded since they are single non-feeder dykes intruded in deep parts of Miocene-Pliocene volcanic edifices; main faults have been taken into account only in those cases where they could modified the superficial movement of magma. The application of kernel density estimation via a linear diffusion process for the volcanic susceptibility assessment has been applied successfully to Lanzarote and could be applied to other fissure volcanic fields worldwide since the results provide information about the probable area where an eruption could take place but also about the main direction of the probable volcanic fissures.
Boundary regularity of Nevanlinna domains and univalent functions in model subspaces
Baranov, Anton D; Fedorovskiy, Konstantin Yu
2011-01-01
In the paper we study boundary regularity of Nevanlinna domains, which have appeared in problems of uniform approximation by polyanalytic polynomials. A new method for constructing Nevanlinna domains with essentially irregular nonanalytic boundaries is suggested; this method is based on finding appropriate univalent functions in model subspaces, that is, in subspaces of the form K Θ =H 2 ominus ΘH 2 , where Θ is an inner function. To describe the irregularity of the boundaries of the domains obtained, recent results by Dolzhenko about boundary regularity of conformal mappings are used. Bibliography: 18 titles.
Embeddings of model subspaces of the Hardy space: compactness and Schatten-von Neumann ideals
Baranov, Anton D
2009-01-01
We study properties of the embedding operators of model subspaces K p Θ (defined by inner functions) in the Hardy space H p (coinvariant subspaces of the shift operator). We find a criterion for the embedding of K p Θ in L p (μ) to be compact similar to the Volberg-Treil theorem on bounded embeddings, and give a positive answer to a question of Cima and Matheson. The proof is based on Bernstein-type inequalities for functions in K p Θ . We investigate measures μ such that the embedding operator belongs to some Schatten-von Neumann ideal.
INDOOR SUBSPACING TO IMPLEMENT INDOORGML FOR INDOOR NAVIGATION
H. Jung
2015-10-01
Full Text Available According to an increasing demand for indoor navigation, there are great attempts to develop applicable indoor network. Representation for a room as a node is not sufficient to apply complex and large buildings. As OGC established IndoorGML, subspacing to partition the space for constructing logical network is introduced. Concerning subspacing for indoor network, transition space like halls or corridors also have to be considered. This study presents the subspacing process for creating an indoor network in shopping mall. Furthermore, categorization of transition space is performed and subspacing of this space is considered. Hall and squares in mall is especially defined for subspacing. Finally, implementation of subspacing process for indoor network is presented.
Indoor Subspacing to Implement Indoorgml for Indoor Navigation
Jung, H.; Lee, J.
2015-10-01
According to an increasing demand for indoor navigation, there are great attempts to develop applicable indoor network. Representation for a room as a node is not sufficient to apply complex and large buildings. As OGC established IndoorGML, subspacing to partition the space for constructing logical network is introduced. Concerning subspacing for indoor network, transition space like halls or corridors also have to be considered. This study presents the subspacing process for creating an indoor network in shopping mall. Furthermore, categorization of transition space is performed and subspacing of this space is considered. Hall and squares in mall is especially defined for subspacing. Finally, implementation of subspacing process for indoor network is presented.
Linear and kernel methods for multi- and hypervariate change detection
Nielsen, Allan Aasbjerg; Canty, Morton J.
2010-01-01
. Principal component analysis (PCA) as well as maximum autocorrelation factor (MAF) and minimum noise fraction (MNF) analyses of IR-MAD images, both linear and kernel-based (which are nonlinear), may further enhance change signals relative to no-change background. The kernel versions are based on a dual...... formulation, also termed Q-mode analysis, in which the data enter into the analysis via inner products in the Gram matrix only. In the kernel version the inner products of the original data are replaced by inner products between nonlinear mappings into higher dimensional feature space. Via kernel substitution......, also known as the kernel trick, these inner products between the mappings are in turn replaced by a kernel function and all quantities needed in the analysis are expressed in terms of the kernel function. This means that we need not know the nonlinear mappings explicitly. Kernel principal component...
Transfer Kernel Common Spatial Patterns for Motor Imagery Brain-Computer Interface Classification
Dai, Mengxi; Liu, Shucong; Zhang, Pengju
2018-01-01
Motor-imagery-based brain-computer interfaces (BCIs) commonly use the common spatial pattern (CSP) as preprocessing step before classification. The CSP method is a supervised algorithm. Therefore a lot of time-consuming training data is needed to build the model. To address this issue, one promising approach is transfer learning, which generalizes a learning model can extract discriminative information from other subjects for target classification task. To this end, we propose a transfer kernel CSP (TKCSP) approach to learn a domain-invariant kernel by directly matching distributions of source subjects and target subjects. The dataset IVa of BCI Competition III is used to demonstrate the validity by our proposed methods. In the experiment, we compare the classification performance of the TKCSP against CSP, CSP for subject-to-subject transfer (CSP SJ-to-SJ), regularizing CSP (RCSP), stationary subspace CSP (ssCSP), multitask CSP (mtCSP), and the combined mtCSP and ssCSP (ss + mtCSP) method. The results indicate that the superior mean classification performance of TKCSP can achieve 81.14%, especially in case of source subjects with fewer number of training samples. Comprehensive experimental evidence on the dataset verifies the effectiveness and efficiency of the proposed TKCSP approach over several state-of-the-art methods. PMID:29743934
On the maximal dimension of a completely entangled subspace for ...
R. Narasimhan (Krishtel eMaging) 1461 1996 Oct 15 13:05:22
dim S = d1d2 ...dk − (d1 +···+ dk) + k − 1, where E is the collection of all completely entangled subspaces. When H1 = H2 and k = 2 an explicit orthonormal basis of a maximal completely entangled subspace of H1 ⊗ H2 is given. We also introduce a more delicate notion of a perfectly entangled subspace for a multipartite ...
Parand, K.; Nikarya, M.
2017-11-01
In this paper a novel method will be introduced to solve a nonlinear partial differential equation (PDE). In the proposed method, we use the spectral collocation method based on Bessel functions of the first kind and the Jacobian free Newton-generalized minimum residual (JFNGMRes) method with adaptive preconditioner. In this work a nonlinear PDE has been converted to a nonlinear system of algebraic equations using the collocation method based on Bessel functions without any linearization, discretization or getting the help of any other methods. Finally, by using JFNGMRes, the solution of the nonlinear algebraic system is achieved. To illustrate the reliability and efficiency of the proposed method, we solve some examples of the famous Fisher equation. We compare our results with other methods.
LogDet Rank Minimization with Application to Subspace Clustering
Zhao Kang
2015-01-01
Full Text Available Low-rank matrix is desired in many machine learning and computer vision problems. Most of the recent studies use the nuclear norm as a convex surrogate of the rank operator. However, all singular values are simply added together by the nuclear norm, and thus the rank may not be well approximated in practical problems. In this paper, we propose using a log-determinant (LogDet function as a smooth and closer, though nonconvex, approximation to rank for obtaining a low-rank representation in subspace clustering. Augmented Lagrange multipliers strategy is applied to iteratively optimize the LogDet-based nonconvex objective function on potentially large-scale data. By making use of the angular information of principal directions of the resultant low-rank representation, an affinity graph matrix is constructed for spectral clustering. Experimental results on motion segmentation and face clustering data demonstrate that the proposed method often outperforms state-of-the-art subspace clustering algorithms.
Bindel, Laurent; Clouet, Laurent; Castanier, Eric; Bonnet, Jerome; Fleury, Guillaume; Vermuse, Manuel; Gamess, Andre; Lejeune, Eric [Societe Generale pour les techniques Nouvelles, Saint Quentin en Yvelines (France)
2000-03-01
The present paper presents the capabilities of a new code named PERCEVAL v4.0 and based on the new point kernel described in an attached issue. Two linked codes named SPECTRE{sub G} and GRAPH{sub 3}D are part of the code package in order to establish the energetic source term and visualize the tri-dimensional scene respectively. (author)
Sparse Event Modeling with Hierarchical Bayesian Kernel Methods
2016-01-05
SECURITY CLASSIFICATION OF: The research objective of this proposal was to develop a predictive Bayesian kernel approach to model count data based on...several predictive variables. Such an approach, which we refer to as the Poisson Bayesian kernel model, is able to model the rate of occurrence of... kernel methods made use of: (i) the Bayesian property of improving predictive accuracy as data are dynamically obtained, and (ii) the kernel function
Stable Kernel Representations as Nonlinear Left Coprime Factorizations
Paice, A.D.B.; Schaft, A.J. van der
1994-01-01
A representation of nonlinear systems based on the idea of representing the input-output pairs of the system as elements of the kernel of a stable operator has been recently introduced. This has been denoted the kernel representation of the system. In this paper it is demonstrated that the kernel
A Fast and Simple Graph Kernel for RDF
de Vries, G.K.D.; de Rooij, S.
2013-01-01
In this paper we study a graph kernel for RDF based on constructing a tree for each instance and counting the number of paths in that tree. In our experiments this kernel shows comparable classification performance to the previously introduced intersection subtree kernel, but is significantly faster
Single pass kernel k-means clustering method
paper proposes a simple and faster version of the kernel k-means clustering ... It has been considered as an important tool ... On the other hand, kernel-based clustering methods, like kernel k-means clus- ..... able at the UCI machine learning repository (Murphy 1994). ... All the data sets have only numeric valued features.
Diah Ratnaningrum
2018-04-01
Full Text Available The suitability of corn kernel as raw material to produce miso fermented by rice-koji containing Aspergillus oryzae and Rhizopus oligosporus has been investigated. The optimization was conducted on two important factors in miso production namely mold composition in rice-koji and salt concentration. The mold composition was prepared by inoculating the spores of 2% A. oryzae, 2% R. oligosporus, and 2% the mixture of both in a ratio of 1:1, 2:1, and 1:2 (v/v into different rice media. The mold composition was optimized to produce rice-koji with high α-amylase and protease activity. Different NaCl concentrations of 10%, 15%, and 20% were subjected to optimization process and added to each mixture after five days of fermentation. The salt concentration was also optimized to produce corn kernel miso with high glucose and high dissolved protein concentration. The result showed that rice-koji containing A. oryzae and R. oligosporus in the ratio of 1:1 had the highest α-amylase and protease activity of 0.42 U/mL and 0.45 U/mL respectively. In addition, the presence of 10% NaCl in corn kernel miso fermented by A. oryzae and R. oligosporus in the ratio of 1:1 exhibited the highest glucose and dissolved protein concentration of 0.64 mg/mL and 8.80 mg/mL respectively. The optimized corn kernel miso by A. oryzae and R. oligosporus in the ratio of 1:1 with 10% NaCl was subjected to nutrient content analysis and compared to the result before the corn kernel was fermented. The nutrient content analysis showed nutrient enhancement after corn kernel was fermented and transformed into a miso. Glucose, dissolved protein, and fat content increased 6.74, 1.34, 7.63 times respectively. This study concludes corn kernel could be utilized to produce a novel corn kernel miso for dietary diversification and for improving nutritional and health status.
Kernel methods in orthogonalization of multi- and hypervariate data
Nielsen, Allan Aasbjerg
2009-01-01
A kernel version of maximum autocorrelation factor (MAF) analysis is described very briefly and applied to change detection in remotely sensed hyperspectral image (HyMap) data. The kernel version is based on a dual formulation also termed Q-mode analysis in which the data enter into the analysis...... via inner products in the Gram matrix only. In the kernel version the inner products are replaced by inner products between nonlinear mappings into higher dimensional feature space of the original data. Via kernel substitution also known as the kernel trick these inner products between the mappings...... are in turn replaced by a kernel function and all quantities needed in the analysis are expressed in terms of this kernel function. This means that we need not know the nonlinear mappings explicitly. Kernel PCA and MAF analysis handle nonlinearities by implicitly transforming data into high (even infinite...
Barndorff-Nielsen, Ole Eiler; Hansen, Peter Reinhard; Lunde, Asger
2011-01-01
In a recent paper we have introduced the class of realised kernel estimators of the increments of quadratic variation in the presence of noise. We showed that this estimator is consistent and derived its limit distribution under various assumptions on the kernel weights. In this paper we extend our...... that subsampling is impotent, in the sense that subsampling has no effect on the asymptotic distribution. Perhaps surprisingly, for the efficient smooth kernels, such as the Parzen kernel, we show that subsampling is harmful as it increases the asymptotic variance. We also study the performance of subsampled...
Ranking Support Vector Machine with Kernel Approximation
Kai Chen
2017-01-01
Full Text Available Learning to rank algorithm has become important in recent years due to its successful application in information retrieval, recommender system, and computational biology, and so forth. Ranking support vector machine (RankSVM is one of the state-of-art ranking models and has been favorably used. Nonlinear RankSVM (RankSVM with nonlinear kernels can give higher accuracy than linear RankSVM (RankSVM with a linear kernel for complex nonlinear ranking problem. However, the learning methods for nonlinear RankSVM are still time-consuming because of the calculation of kernel matrix. In this paper, we propose a fast ranking algorithm based on kernel approximation to avoid computing the kernel matrix. We explore two types of kernel approximation methods, namely, the Nyström method and random Fourier features. Primal truncated Newton method is used to optimize the pairwise L2-loss (squared Hinge-loss objective function of the ranking model after the nonlinear kernel approximation. Experimental results demonstrate that our proposed method gets a much faster training speed than kernel RankSVM and achieves comparable or better performance over state-of-the-art ranking algorithms.
Ranking Support Vector Machine with Kernel Approximation.
Chen, Kai; Li, Rongchun; Dou, Yong; Liang, Zhengfa; Lv, Qi
2017-01-01
Learning to rank algorithm has become important in recent years due to its successful application in information retrieval, recommender system, and computational biology, and so forth. Ranking support vector machine (RankSVM) is one of the state-of-art ranking models and has been favorably used. Nonlinear RankSVM (RankSVM with nonlinear kernels) can give higher accuracy than linear RankSVM (RankSVM with a linear kernel) for complex nonlinear ranking problem. However, the learning methods for nonlinear RankSVM are still time-consuming because of the calculation of kernel matrix. In this paper, we propose a fast ranking algorithm based on kernel approximation to avoid computing the kernel matrix. We explore two types of kernel approximation methods, namely, the Nyström method and random Fourier features. Primal truncated Newton method is used to optimize the pairwise L2-loss (squared Hinge-loss) objective function of the ranking model after the nonlinear kernel approximation. Experimental results demonstrate that our proposed method gets a much faster training speed than kernel RankSVM and achieves comparable or better performance over state-of-the-art ranking algorithms.
Metabolic network prediction through pairwise rational kernels.
Roche-Lima, Abiel; Domaratzki, Michael; Fristensky, Brian
2014-09-26
Metabolic networks are represented by the set of metabolic pathways. Metabolic pathways are a series of biochemical reactions, in which the product (output) from one reaction serves as the substrate (input) to another reaction. Many pathways remain incompletely characterized. One of the major challenges of computational biology is to obtain better models of metabolic pathways. Existing models are dependent on the annotation of the genes. This propagates error accumulation when the pathways are predicted by incorrectly annotated genes. Pairwise classification methods are supervised learning methods used to classify new pair of entities. Some of these classification methods, e.g., Pairwise Support Vector Machines (SVMs), use pairwise kernels. Pairwise kernels describe similarity measures between two pairs of entities. Using pairwise kernels to handle sequence data requires long processing times and large storage. Rational kernels are kernels based on weighted finite-state transducers that represent similarity measures between sequences or automata. They have been effectively used in problems that handle large amount of sequence information such as protein essentiality, natural language processing and machine translations. We create a new family of pairwise kernels using weighted finite-state transducers (called Pairwise Rational Kernel (PRK)) to predict metabolic pathways from a variety of biological data. PRKs take advantage of the simpler representations and faster algorithms of transducers. Because raw sequence data can be used, the predictor model avoids the errors introduced by incorrect gene annotations. We then developed several experiments with PRKs and Pairwise SVM to validate our methods using the metabolic network of Saccharomyces cerevisiae. As a result, when PRKs are used, our method executes faster in comparison with other pairwise kernels. Also, when we use PRKs combined with other simple kernels that include evolutionary information, the accuracy
A block Krylov subspace time-exact solution method for linear ordinary differential equation systems
Bochev, Mikhail A.
2013-01-01
We propose a time-exact Krylov-subspace-based method for solving linear ordinary differential equation systems of the form $y'=-Ay+g(t)$ and $y"=-Ay+g(t)$, where $y(t)$ is the unknown function. The method consists of two stages. The first stage is an accurate piecewise polynomial approximation of
Tamiminia, Haifa; Homayouni, Saeid; McNairn, Heather; Safari, Abdoreza
2017-06-01
Polarimetric Synthetic Aperture Radar (PolSAR) data, thanks to their specific characteristics such as high resolution, weather and daylight independence, have become a valuable source of information for environment monitoring and management. The discrimination capability of observations acquired by these sensors can be used for land cover classification and mapping. The aim of this paper is to propose an optimized kernel-based C-means clustering algorithm for agriculture crop mapping from multi-temporal PolSAR data. Firstly, several polarimetric features are extracted from preprocessed data. These features are linear polarization intensities, and several statistical and physical based decompositions such as Cloude-Pottier, Freeman-Durden and Yamaguchi techniques. Then, the kernelized version of hard and fuzzy C-means clustering algorithms are applied to these polarimetric features in order to identify crop types. The kernel function, unlike the conventional partitioning clustering algorithms, simplifies the non-spherical and non-linearly patterns of data structure, to be clustered easily. In addition, in order to enhance the results, Particle Swarm Optimization (PSO) algorithm is used to tune the kernel parameters, cluster centers and to optimize features selection. The efficiency of this method was evaluated by using multi-temporal UAVSAR L-band images acquired over an agricultural area near Winnipeg, Manitoba, Canada, during June and July in 2012. The results demonstrate more accurate crop maps using the proposed method when compared to the classical approaches, (e.g. 12% improvement in general). In addition, when the optimization technique is used, greater improvement is observed in crop classification, e.g. 5% in overall. Furthermore, a strong relationship between Freeman-Durden volume scattering component, which is related to canopy structure, and phenological growth stages is observed.
Lestari, A. W.; Rustam, Z.
2017-07-01
In the last decade, breast cancer has become the focus of world attention as this disease is one of the primary leading cause of death for women. Therefore, it is necessary to have the correct precautions and treatment. In previous studies, Fuzzy Kennel K-Medoid algorithm has been used for multi-class data. This paper proposes an algorithm to classify the high dimensional data of breast cancer using Fuzzy Possibilistic C-means (FPCM) and a new method based on clustering analysis using Normed Kernel Function-Based Fuzzy Possibilistic C-Means (NKFPCM). The objective of this paper is to obtain the best accuracy in classification of breast cancer data. In order to improve the accuracy of the two methods, the features candidates are evaluated using feature selection, where Laplacian Score is used. The results show the comparison accuracy and running time of FPCM and NKFPCM with and without feature selection.
High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D
Wittmann, Roland; Bungartz, Hans-Joachim; Neumann, Philipp
2017-01-01
-by-step transformation of the second order finite volume scheme in FullSWOF2D towards MPI parallelization. Second, the computational kernels are optimized by the use of templates and a portable vectorization approach. We discuss the load imbalance of the flux computation
KIRMES: kernel-based identification of regulatory modules in euchromatic sequences.
Schultheiss, Sebastian J; Busch, Wolfgang; Lohmann, Jan U; Kohlbacher, Oliver; Rätsch, Gunnar
2009-08-15
Understanding transcriptional regulation is one of the main challenges in computational biology. An important problem is the identification of transcription factor (TF) binding sites in promoter regions of potential TF target genes. It is typically approached by position weight matrix-based motif identification algorithms using Gibbs sampling, or heuristics to extend seed oligos. Such algorithms succeed in identifying single, relatively well-conserved binding sites, but tend to fail when it comes to the identification of combinations of several degenerate binding sites, as those often found in cis-regulatory modules. We propose a new algorithm that combines the benefits of existing motif finding with the ones of support vector machines (SVMs) to find degenerate motifs in order to improve the modeling of regulatory modules. In experiments on microarray data from Arabidopsis thaliana, we were able to show that the newly developed strategy significantly improves the recognition of TF targets. The python source code (open source-licensed under GPL), the data for the experiments and a Galaxy-based web service are available at http://www.fml.mpg.de/raetsch/suppl/kirmes/.
Robust Building Energy Load Forecasting Using Physically-Based Kernel Models
Anand Krishnan Prakash
2018-04-01
Full Text Available Robust and accurate building energy load forecasting is important for helping building managers and utilities to plan, budget, and strategize energy resources in advance. With recent prevalent adoption of smart-meters in buildings, a significant amount of building energy consumption data became available. Many studies have developed physics-based white box models and data-driven black box models to predict building energy consumption; however, they require extensive prior knowledge about building system, need a large set of training data, or lack robustness to different forecasting scenarios. In this paper, we introduce a new building energy forecasting method based on Gaussian Process Regression (GPR that incorporates physical insights about load data characteristics to improve accuracy while reducing training requirements. The GPR is a non-parametric regression method that models the data as a joint Gaussian distribution with mean and covariance functions and forecast using the Bayesian updating. We model the covariance function of the GPR to reflect the data patterns in different forecasting horizon scenarios, as prior knowledge. Our method takes advantage of the modeling flexibility and computational efficiency of the GPR while benefiting from the physical insights to further improve the training efficiency and accuracy. We evaluate our method with three field datasets from two university campuses (Carnegie Mellon University and Stanford University for both short- and long-term load forecasting. The results show that our method performs more accurately, especially when the training dataset is small, compared to other state-of-the-art forecasting models (up to 2.95 times smaller prediction error.
Chen Shih-Wei
2011-11-01
Full Text Available Abstract Background The computer-aided identification of specific gait patterns is an important issue in the assessment of Parkinson's disease (PD. In this study, a computer vision-based gait analysis approach is developed to assist the clinical assessments of PD with kernel-based principal component analysis (KPCA. Method Twelve PD patients and twelve healthy adults with no neurological history or motor disorders within the past six months were recruited and separated according to their "Non-PD", "Drug-On", and "Drug-Off" states. The participants were asked to wear light-colored clothing and perform three walking trials through a corridor decorated with a navy curtain at their natural pace. The participants' gait performance during the steady-state walking period was captured by a digital camera for gait analysis. The collected walking image frames were then transformed into binary silhouettes for noise reduction and compression. Using the developed KPCA-based method, the features within the binary silhouettes can be extracted to quantitatively determine the gait cycle time, stride length, walking velocity, and cadence. Results and Discussion The KPCA-based method uses a feature-extraction approach, which was verified to be more effective than traditional image area and principal component analysis (PCA approaches in classifying "Non-PD" controls and "Drug-Off/On" PD patients. Encouragingly, this method has a high accuracy rate, 80.51%, for recognizing different gaits. Quantitative gait parameters are obtained, and the power spectrums of the patients' gaits are analyzed. We show that that the slow and irregular actions of PD patients during walking tend to transfer some of the power from the main lobe frequency to a lower frequency band. Our results indicate the feasibility of using gait performance to evaluate the motor function of patients with PD. Conclusion This KPCA-based method requires only a digital camera and a decorated corridor setup
A noise reduction technique based on nonlinear kernel function for heart sound analysis.
Mondal, Ashok; Saxena, Ishan; Tang, Hong; Banerjee, Poulami
2017-02-13
The main difficulty encountered in interpretation of cardiac sound is interference of noise. The contaminated noise obscures the relevant information which are useful for recognition of heart diseases. The unwanted signals are produced mainly by lungs and surrounding environment. In this paper, a novel heart sound de-noising technique has been introduced based on a combined framework of wavelet packet transform (WPT) and singular value decomposition (SVD). The most informative node of wavelet tree is selected on the criteria of mutual information measurement. Next, the coefficient corresponding to the selected node is processed by SVD technique to suppress noisy component from heart sound signal. To justify the efficacy of the proposed technique, several experiments have been conducted with heart sound dataset, including normal and pathological cases at different signal to noise ratios. The significance of the method is validated by statistical analysis of the results. The biological information preserved in de-noised heart sound (HS) signal is evaluated by k-means clustering algorithm and Fit Factor calculation. The overall results show that proposed method is superior than the baseline methods.
iTRAQ-Based Proteomics Analysis and Network Integration for Kernel Tissue Development in Maize
Dong, Yongbin; Wang, Qilei; Du, Chunguang; Xiong, Wenwei; Li, Xinyu; Zhu, Sailan; Li, Yuling
2017-01-01
Grain weight is one of the most important yield components and a developmentally complex structure comprised of two major compartments (endosperm and pericarp) in maize (Zea mays L.), however, very little is known concerning the coordinated accumulation of the numerous proteins involved. Herein, we used isobaric tags for relative and absolute quantitation (iTRAQ)-based comparative proteomic method to analyze the characteristics of dynamic proteomics for endosperm and pericarp during grain development. Totally, 9539 proteins were identified for both components at four development stages, among which 1401 proteins were non-redundant, 232 proteins were specific in pericarp and 153 proteins were specific in endosperm. A functional annotation of the identified proteins revealed the importance of metabolic and cellular processes, and binding and catalytic activities for the tissue development. Three and 76 proteins involved in 49 Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were integrated for the specific endosperm and pericarp proteins, respectively, reflecting their complex metabolic interactions. In addition, four proteins with important functions and different expression levels were chosen for gene cloning and expression analysis. Different concordance between mRNA level and the protein abundance was observed across different proteins, stages, and tissues as in previous research. These results could provide useful message for understanding the developmental mechanisms in grain development in maize. PMID:28837076
Duff, I.
1994-12-31
This workshop focuses on kernels for iterative software packages. Specifically, the three speakers discuss various aspects of sparse BLAS kernels. Their topics are: `Current status of user lever sparse BLAS`; Current status of the sparse BLAS toolkit`; and `Adding matrix-matrix and matrix-matrix-matrix multiply to the sparse BLAS toolkit`.
On the numerical stability analysis of pipelined Krylov subspace methods
Carson, E.T.; Rozložník, Miroslav; Strakoš, Z.; Tichý, P.; Tůma, M.
submitted 2017 (2018) R&D Projects: GA ČR GA13-06684S Grant - others:GA MŠk(CZ) LL1202 Institutional support: RVO:67985807 Keywords : Krylov subspace methods * the conjugate gradient method * numerical stability * inexact computations * delay of convergence * maximal attainable accuracy * pipelined Krylov subspace methods * exascale computations
Subspace methods for pattern recognition in intelligent environment
Jain, Lakhmi
2014-01-01
This research book provides a comprehensive overview of the state-of-the-art subspace learning methods for pattern recognition in intelligent environment. With the fast development of internet and computer technologies, the amount of available data is rapidly increasing in our daily life. How to extract core information or useful features is an important issue. Subspace methods are widely used for dimension reduction and feature extraction in pattern recognition. They transform a high-dimensional data to a lower-dimensional space (subspace), where most information is retained. The book covers a broad spectrum of subspace methods including linear, nonlinear and multilinear subspace learning methods and applications. The applications include face alignment, face recognition, medical image analysis, remote sensing image classification, traffic sign recognition, image clustering, super resolution, edge detection, multi-view facial image synthesis.
Kim, Song Hyun; Woo, Myeong Hyun; Shin, Chang Ho; Pyeon, Cheol Ho
2015-01-01
In this study, a new balance equation to overcome the problems generated by the previous methods is proposed using source-based balance equation. And then, a simple problem is analyzed with the proposed method. In this study, a source-based balance equation with the time dependent fission kernel was derived to simplify the kinetics equation. To analyze the partial variations of reactor characteristics, two representative methods were introduced in previous studies; (1) quasi-statics method and (2) multipoint technique. The main idea of quasistatics method is to use a low-order approximation for large integration times. To realize the quasi-statics method, first, time dependent flux is separated into the shape and amplitude functions, and shape function is calculated. It is noted that the method has a good accuracy; however, it can be expensive as a calculation cost aspect because the shape function should be fully recalculated to obtain accurate results. To improve the calculation efficiency, multipoint method was proposed. The multipoint method is based on the classic kinetics equation with using Green's function to analyze the flight probability from region r' to r. Those previous methods have been used to analyze the reactor kinetics analysis; however, the previous methods can have some limitations. First, three group variables (r g , E g , t g ) should be considered to solve the time dependent balance equation. This leads a big limitation to apply large system problem with good accuracy. Second, the energy group neutrons should be used to analyze reactor kinetics problems. In time dependent problem, neutron energy distribution can be changed at different time. It can affect the change of the group cross section; therefore, it can lead the accuracy problem. Third, the neutrons in a space-time region continually affect the other space-time regions; however, it is not properly considered in the previous method. Using birth history of the neutron sources
Kim, Song Hyun; Woo, Myeong Hyun; Shin, Chang Ho [Hanyang University, Seoul (Korea, Republic of); Pyeon, Cheol Ho [Kyoto University, Osaka (Japan)
2015-10-15
In this study, a new balance equation to overcome the problems generated by the previous methods is proposed using source-based balance equation. And then, a simple problem is analyzed with the proposed method. In this study, a source-based balance equation with the time dependent fission kernel was derived to simplify the kinetics equation. To analyze the partial variations of reactor characteristics, two representative methods were introduced in previous studies; (1) quasi-statics method and (2) multipoint technique. The main idea of quasistatics method is to use a low-order approximation for large integration times. To realize the quasi-statics method, first, time dependent flux is separated into the shape and amplitude functions, and shape function is calculated. It is noted that the method has a good accuracy; however, it can be expensive as a calculation cost aspect because the shape function should be fully recalculated to obtain accurate results. To improve the calculation efficiency, multipoint method was proposed. The multipoint method is based on the classic kinetics equation with using Green's function to analyze the flight probability from region r' to r. Those previous methods have been used to analyze the reactor kinetics analysis; however, the previous methods can have some limitations. First, three group variables (r{sub g}, E{sub g}, t{sub g}) should be considered to solve the time dependent balance equation. This leads a big limitation to apply large system problem with good accuracy. Second, the energy group neutrons should be used to analyze reactor kinetics problems. In time dependent problem, neutron energy distribution can be changed at different time. It can affect the change of the group cross section; therefore, it can lead the accuracy problem. Third, the neutrons in a space-time region continually affect the other space-time regions; however, it is not properly considered in the previous method. Using birth history of the
Ulf Hensen
Full Text Available We develop a general minimally coupled subspace approach (MCSA to compute absolute entropies of macromolecules, such as proteins, from computer generated canonical ensembles. Our approach overcomes limitations of current estimates such as the quasi-harmonic approximation which neglects non-linear and higher-order correlations as well as multi-minima characteristics of protein energy landscapes. Here, Full Correlation Analysis, adaptive kernel density estimation, and mutual information expansions are combined and high accuracy is demonstrated for a number of test systems ranging from alkanes to a 14 residue peptide. We further computed the configurational entropy for the full 67-residue cofactor of the TATA box binding protein illustrating that MCSA yields improved results also for large macromolecular systems.
Fonville, Judith M.; Bylesjoe, Max; Coen, Muireann; Nicholson, Jeremy K.; Holmes, Elaine; Lindon, John C.; Rantalainen, Mattias
2011-01-01
Highlights: → Non-linear modeling of metabonomic data using K-OPLS. → automated optimization of the kernel parameter by simulated annealing. → K-OPLS provides improved prediction performance for exemplar spectral data sets. → software implementation available for R and Matlab under GPL v2 license. - Abstract: Linear multivariate projection methods are frequently applied for predictive modeling of spectroscopic data in metabonomic studies. The OPLS method is a commonly used computational procedure for characterizing spectral metabonomic data, largely due to its favorable model interpretation properties providing separate descriptions of predictive variation and response-orthogonal structured noise. However, when the relationship between descriptor variables and the response is non-linear, conventional linear models will perform sub-optimally. In this study we have evaluated to what extent a non-linear model, kernel-based orthogonal projections to latent structures (K-OPLS), can provide enhanced predictive performance compared to the linear OPLS model. Just like its linear counterpart, K-OPLS provides separate model components for predictive variation and response-orthogonal structured noise. The improved model interpretation by this separate modeling is a property unique to K-OPLS in comparison to other kernel-based models. Simulated annealing (SA) was used for effective and automated optimization of the kernel-function parameter in K-OPLS (SA-K-OPLS). Our results reveal that the non-linear K-OPLS model provides improved prediction performance in three separate metabonomic data sets compared to the linear OPLS model. We also demonstrate how response-orthogonal K-OPLS components provide valuable biological interpretation of model and data. The metabonomic data sets were acquired using proton Nuclear Magnetic Resonance (NMR) spectroscopy, and include a study of the liver toxin galactosamine, a study of the nephrotoxin mercuric chloride and a study of
Putting Priors in Mixture Density Mercer Kernels
Srivastava, Ashok N.; Schumann, Johann; Fischer, Bernd
2004-01-01
This paper presents a new methodology for automatic knowledge driven data mining based on the theory of Mercer Kernels, which are highly nonlinear symmetric positive definite mappings from the original image space to a very high, possibly infinite dimensional feature space. We describe a new method called Mixture Density Mercer Kernels to learn kernel function directly from data, rather than using predefined kernels. These data adaptive kernels can en- code prior knowledge in the kernel using a Bayesian formulation, thus allowing for physical information to be encoded in the model. We compare the results with existing algorithms on data from the Sloan Digital Sky Survey (SDSS). The code for these experiments has been generated with the AUTOBAYES tool, which automatically generates efficient and documented C/C++ code from abstract statistical model specifications. The core of the system is a schema library which contains template for learning and knowledge discovery algorithms like different versions of EM, or numeric optimization methods like conjugate gradient methods. The template instantiation is supported by symbolic- algebraic computations, which allows AUTOBAYES to find closed-form solutions and, where possible, to integrate them into the code. The results show that the Mixture Density Mercer-Kernel described here outperforms tree-based classification in distinguishing high-redshift galaxies from low- redshift galaxies by approximately 16% on test data, bagged trees by approximately 7%, and bagged trees built on a much larger sample of data by approximately 2%.
Ensemble Approach to Building Mercer Kernels
National Aeronautics and Space Administration — This paper presents a new methodology for automatic knowledge driven data mining based on the theory of Mercer Kernels, which are highly nonlinear symmetric positive...
Generalization Performance of Regularized Ranking With Multiscale Kernels.
Zhou, Yicong; Chen, Hong; Lan, Rushi; Pan, Zhibin
2016-05-01
The regularized kernel method for the ranking problem has attracted increasing attentions in machine learning. The previous regularized ranking algorithms are usually based on reproducing kernel Hilbert spaces with a single kernel. In this paper, we go beyond this framework by investigating the generalization performance of the regularized ranking with multiscale kernels. A novel ranking algorithm with multiscale kernels is proposed and its representer theorem is proved. We establish the upper bound of the generalization error in terms of the complexity of hypothesis spaces. It shows that the multiscale ranking algorithm can achieve satisfactory learning rates under mild conditions. Experiments demonstrate the effectiveness of the proposed method for drug discovery and recommendation tasks.
Sparse subspace clustering for data with missing entries and high-rank matrix completion.
Fan, Jicong; Chow, Tommy W S
2017-09-01
Many methods have recently been proposed for subspace clustering, but they are often unable to handle incomplete data because of missing entries. Using matrix completion methods to recover missing entries is a common way to solve the problem. Conventional matrix completion methods require that the matrix should be of low-rank intrinsically, but most matrices are of high-rank or even full-rank in practice, especially when the number of subspaces is large. In this paper, a new method called Sparse Representation with Missing Entries and Matrix Completion is proposed to solve the problems of incomplete-data subspace clustering and high-rank matrix completion. The proposed algorithm alternately computes the matrix of sparse representation coefficients and recovers the missing entries of a data matrix. The proposed algorithm recovers missing entries through minimizing the representation coefficients, representation errors, and matrix rank. Thorough experimental study and comparative analysis based on synthetic data and natural images were conducted. The presented results demonstrate that the proposed algorithm is more effective in subspace clustering and matrix completion compared with other existing methods. Copyright © 2017 Elsevier Ltd. All rights reserved.
Classification With Truncated Distance Kernel.
Huang, Xiaolin; Suykens, Johan A K; Wang, Shuning; Hornegger, Joachim; Maier, Andreas
2018-05-01
This brief proposes a truncated distance (TL1) kernel, which results in a classifier that is nonlinear in the global region but is linear in each subregion. With this kernel, the subregion structure can be trained using all the training data and local linear classifiers can be established simultaneously. The TL1 kernel has good adaptiveness to nonlinearity and is suitable for problems which require different nonlinearities in different areas. Though the TL1 kernel is not positive semidefinite, some classical kernel learning methods are still applicable which means that the TL1 kernel can be directly used in standard toolboxes by replacing the kernel evaluation. In numerical experiments, the TL1 kernel with a pregiven parameter achieves similar or better performance than the radial basis function kernel with the parameter tuned by cross validation, implying the TL1 kernel a promising nonlinear kernel for classification tasks.
Linear Subspace Ranking Hashing for Cross-Modal Retrieval.
Li, Kai; Qi, Guo-Jun; Ye, Jun; Hua, Kien A
2017-09-01
Hashing has attracted a great deal of research in recent years due to its effectiveness for the retrieval and indexing of large-scale high-dimensional multimedia data. In this paper, we propose a novel ranking-based hashing framework that maps data from different modalities into a common Hamming space where the cross-modal similarity can be measured using Hamming distance. Unlike existing cross-modal hashing algorithms where the learned hash functions are binary space partitioning functions, such as the sign and threshold function, the proposed hashing scheme takes advantage of a new class of hash functions closely related to rank correlation measures which are known to be scale-invariant, numerically stable, and highly nonlinear. Specifically, we jointly learn two groups of linear subspaces, one for each modality, so that features' ranking orders in different linear subspaces maximally preserve the cross-modal similarities. We show that the ranking-based hash function has a natural probabilistic approximation which transforms the original highly discontinuous optimization problem into one that can be efficiently solved using simple gradient descent algorithms. The proposed hashing framework is also flexible in the sense that the optimization procedures are not tied up to any specific form of loss function, which is typical for existing cross-modal hashing methods, but rather we can flexibly accommodate different loss functions with minimal changes to the learning steps. We demonstrate through extensive experiments on four widely-used real-world multimodal datasets that the proposed cross-modal hashing method can achieve competitive performance against several state-of-the-arts with only moderate training and testing time.
Robust adaptive subspace detection in impulsive noise
Ben Atitallah, Ismail
2016-09-13
This paper addresses the design of the Adaptive Subspace Matched Filter (ASMF) detector in the presence of compound Gaussian clutters and a mismatch in the steering vector. In particular, we consider the case wherein the ASMF uses the regularized Tyler estimator (RTE) to estimate the clutter covariance matrix. Under this setting, a major question that needs to be addressed concerns the setting of the threshold and the regularization parameter. To answer this question, we consider the regime in which the number of observations used to estimate the RTE and their dimensions grow large together. Recent results from random matrix theory are then used in order to approximate the false alarm and detection probabilities by deterministic quantities. The latter are optimized in order to maximize an upper bound on the asymptotic detection probability while keeping the asymptotic false alarm probability at a fixed rate. © 2016 IEEE.
Robust adaptive subspace detection in impulsive noise
Ben Atitallah, Ismail; Kammoun, Abla; Alouini, Mohamed-Slim; Al-Naffouri, Tareq Y.
2016-01-01
This paper addresses the design of the Adaptive Subspace Matched Filter (ASMF) detector in the presence of compound Gaussian clutters and a mismatch in the steering vector. In particular, we consider the case wherein the ASMF uses the regularized Tyler estimator (RTE) to estimate the clutter covariance matrix. Under this setting, a major question that needs to be addressed concerns the setting of the threshold and the regularization parameter. To answer this question, we consider the regime in which the number of observations used to estimate the RTE and their dimensions grow large together. Recent results from random matrix theory are then used in order to approximate the false alarm and detection probabilities by deterministic quantities. The latter are optimized in order to maximize an upper bound on the asymptotic detection probability while keeping the asymptotic false alarm probability at a fixed rate. © 2016 IEEE.
Gärtner, Thomas
2009-01-01
This book provides a unique treatment of an important area of machine learning and answers the question of how kernel methods can be applied to structured data. Kernel methods are a class of state-of-the-art learning algorithms that exhibit excellent learning results in several application domains. Originally, kernel methods were developed with data in mind that can easily be embedded in a Euclidean vector space. Much real-world data does not have this property but is inherently structured. An example of such data, often consulted in the book, is the (2D) graph structure of molecules formed by
Exact Heat Kernel on a Hypersphere and Its Applications in Kernel SVM
Chenchao Zhao
2018-01-01
Full Text Available Many contemporary statistical learning methods assume a Euclidean feature space. This paper presents a method for defining similarity based on hyperspherical geometry and shows that it often improves the performance of support vector machine compared to other competing similarity measures. Specifically, the idea of using heat diffusion on a hypersphere to measure similarity has been previously proposed and tested by Lafferty and Lebanon [1], demonstrating promising results based on a heuristic heat kernel obtained from the zeroth order parametrix expansion; however, how well this heuristic kernel agrees with the exact hyperspherical heat kernel remains unknown. This paper presents a higher order parametrix expansion of the heat kernel on a unit hypersphere and discusses several problems associated with this expansion method. We then compare the heuristic kernel with an exact form of the heat kernel expressed in terms of a uniformly and absolutely convergent series in high-dimensional angular momentum eigenmodes. Being a natural measure of similarity between sample points dwelling on a hypersphere, the exact kernel often shows superior performance in kernel SVM classifications applied to text mining, tumor somatic mutation imputation, and stock market analysis.
An alternative subspace approach to EEG dipole source localization
Xu, Xiao-Liang; Xu, Bobby; He, Bin
2004-01-01
In the present study, we investigate a new approach to electroencephalography (EEG) three-dimensional (3D) dipole source localization by using a non-recursive subspace algorithm called FINES. In estimating source dipole locations, the present approach employs projections onto a subspace spanned by a small set of particular vectors (FINES vector set) in the estimated noise-only subspace instead of the entire estimated noise-only subspace in the case of classic MUSIC. The subspace spanned by this vector set is, in the sense of principal angle, closest to the subspace spanned by the array manifold associated with a particular brain region. By incorporating knowledge of the array manifold in identifying FINES vector sets in the estimated noise-only subspace for different brain regions, the present approach is able to estimate sources with enhanced accuracy and spatial resolution, thus enhancing the capability of resolving closely spaced sources and reducing estimation errors. The present computer simulations show, in EEG 3D dipole source localization, that compared to classic MUSIC, FINES has (1) better resolvability of two closely spaced dipolar sources and (2) better estimation accuracy of source locations. In comparison with RAP-MUSIC, FINES' performance is also better for the cases studied when the noise level is high and/or correlations among dipole sources exist.
An alternative subspace approach to EEG dipole source localization
Xu Xiaoliang; Xu, Bobby; He Bin
2004-01-01
In the present study, we investigate a new approach to electroencephalography (EEG) three-dimensional (3D) dipole source localization by using a non-recursive subspace algorithm called FINES. In estimating source dipole locations, the present approach employs projections onto a subspace spanned by a small set of particular vectors (FINES vector set) in the estimated noise-only subspace instead of the entire estimated noise-only subspace in the case of classic MUSIC. The subspace spanned by this vector set is, in the sense of principal angle, closest to the subspace spanned by the array manifold associated with a particular brain region. By incorporating knowledge of the array manifold in identifying FINES vector sets in the estimated noise-only subspace for different brain regions, the present approach is able to estimate sources with enhanced accuracy and spatial resolution, thus enhancing the capability of resolving closely spaced sources and reducing estimation errors. The present computer simulations show, in EEG 3D dipole source localization, that compared to classic MUSIC, FINES has (1) better resolvability of two closely spaced dipolar sources and (2) better estimation accuracy of source locations. In comparison with RAP-MUSIC, FINES' performance is also better for the cases studied when the noise level is high and/or correlations among dipole sources exist
Locally linear approximation for Kernel methods : the Railway Kernel
Muñoz, Alberto; González, Javier
2008-01-01
In this paper we present a new kernel, the Railway Kernel, that works properly for general (nonlinear) classification problems, with the interesting property that acts locally as a linear kernel. In this way, we avoid potential problems due to the use of a general purpose kernel, like the RBF kernel, as the high dimension of the induced feature space. As a consequence, following our methodology the number of support vectors is much lower and, therefore, the generalization capab...
Lyapunov vectors and assimilation in the unstable subspace: theory and applications
Palatella, Luigi; Carrassi, Alberto; Trevisan, Anna
2013-01-01
Based on a limited number of noisy observations, estimation algorithms provide a complete description of the state of a system at current time. Estimation algorithms that go under the name of assimilation in the unstable subspace (AUS) exploit the nonlinear stability properties of the forecasting model in their formulation. Errors that grow due to sensitivity to initial conditions are efficiently removed by confining the analysis solution in the unstable and neutral subspace of the system, the subspace spanned by Lyapunov vectors with positive and zero exponents, while the observational noise does not disturb the system along the stable directions. The formulation of the AUS approach in the context of four-dimensional variational assimilation (4DVar-AUS) and the extended Kalman filter (EKF-AUS) and its application to chaotic models is reviewed. In both instances, the AUS algorithms are at least as efficient but simpler to implement and computationally less demanding than their original counterparts. As predicted by the theory when error dynamics is linear, the optimal subspace dimension for 4DVar-AUS is given by the number of positive and null Lyapunov exponents, while the EKF-AUS algorithm, using the same unstable and neutral subspace, recovers the solution of the full EKF algorithm, but dealing with error covariance matrices of a much smaller dimension and significantly reducing the computational burden. Examples of the application to a simplified model of the atmospheric circulation and to the optimal velocity model for traffic dynamics are given. This article is part of a special issue of Journal of Physics A: Mathematical and Theoretical devoted to ‘Lyapunov analysis: from dynamical systems theory to applications’. (paper)
Motai, Yuichi
2015-01-01
Describes and discusses the variants of kernel analysis methods for data types that have been intensely studied in recent years This book covers kernel analysis topics ranging from the fundamental theory of kernel functions to its applications. The book surveys the current status, popular trends, and developments in kernel analysis studies. The author discusses multiple kernel learning algorithms and how to choose the appropriate kernels during the learning phase. Data-Variant Kernel Analysis is a new pattern analysis framework for different types of data configurations. The chapters include
Zhou, Ping; Song, Heda; Wang, Hong; Chai, Tianyou
2017-09-01
Blast furnace (BF) in ironmaking is a nonlinear dynamic process with complicated physical-chemical reactions, where multi-phase and multi-field coupling and large time delay occur during its operation. In BF operation, the molten iron temperature (MIT) as well as Si, P and S contents of molten iron are the most essential molten iron quality (MIQ) indices, whose measurement, modeling and control have always been important issues in metallurgic engineering and automation field. This paper develops a novel data-driven nonlinear state space modeling for the prediction and control of multivariate MIQ indices by integrating hybrid modeling and control techniques. First, to improve modeling efficiency, a data-driven hybrid method combining canonical correlation analysis and correlation analysis is proposed to identify the most influential controllable variables as the modeling inputs from multitudinous factors would affect the MIQ indices. Then, a Hammerstein model for the prediction of MIQ indices is established using the LS-SVM based nonlinear subspace identification method. Such a model is further simplified by using piecewise cubic Hermite interpolating polynomial method to fit the complex nonlinear kernel function. Compared to the original Hammerstein model, this simplified model can not only significantly reduce the computational complexity, but also has almost the same reliability and accuracy for a stable prediction of MIQ indices. Last, in order to verify the practicability of the developed model, it is applied in designing a genetic algorithm based nonlinear predictive controller for multivariate MIQ indices by directly taking the established model as a predictor. Industrial experiments show the advantages and effectiveness of the proposed approach.
Microscopic theory of dynamical subspace for large amplitude collective motion
Sakata, Fumihiko; Marumori, Toshio; Ogura, Masanori.
1986-01-01
A full quantum theory appropriate for describing large amplitude collective motion is proposed by exploiting the basic idea of the semi-classical theory so far developed within the time-depedent Hartree-Fock theory. A central problem of the quantum theory is how to determine an optimal representation called a dynamical representation specific for the collective subspace where the large amplitude collective motion is replicated as precisely as possible. As an extension of the semi-classical theory where the concept of an approximate integral surface played an important role, the collective subspace is properly characterized by introducing a concept of an approximate invariant subspace of the Hamiltonian. (author)
A New Inexact Inverse Subspace Iteration for Generalized Eigenvalue Problems
Fatemeh Mohammad
2014-05-01
Full Text Available In this paper, we represent an inexact inverse subspace iteration method for computing a few eigenpairs of the generalized eigenvalue problem $Ax = \\lambda Bx$[Q.~Ye and P.~Zhang, Inexact inverse subspace iteration for generalized eigenvalue problems, Linear Algebra and its Application, 434 (2011 1697-1715]. In particular, the linear convergence property of the inverse subspace iteration is preserved.
Kernel Bayesian ART and ARTMAP.
Masuyama, Naoki; Loo, Chu Kiong; Dawood, Farhan
2018-02-01
Adaptive Resonance Theory (ART) is one of the successful approaches to resolving "the plasticity-stability dilemma" in neural networks, and its supervised learning model called ARTMAP is a powerful tool for classification. Among several improvements, such as Fuzzy or Gaussian based models, the state of art model is Bayesian based one, while solving the drawbacks of others. However, it is known that the Bayesian approach for the high dimensional and a large number of data requires high computational cost, and the covariance matrix in likelihood becomes unstable. This paper introduces Kernel Bayesian ART (KBA) and ARTMAP (KBAM) by integrating Kernel Bayes' Rule (KBR) and Correntropy Induced Metric (CIM) to Bayesian ART (BA) and ARTMAP (BAM), respectively, while maintaining the properties of BA and BAM. The kernel frameworks in KBA and KBAM are able to avoid the curse of dimensionality. In addition, the covariance-free Bayesian computation by KBR provides the efficient and stable computational capability to KBA and KBAM. Furthermore, Correntropy-based similarity measurement allows improving the noise reduction ability even in the high dimensional space. The simulation experiments show that KBA performs an outstanding self-organizing capability than BA, and KBAM provides the superior classification ability than BAM, respectively. Copyright © 2017 Elsevier Ltd. All rights reserved.
do Nascimento Silva, Jaqueline; Mascarin, Gabriel Moura; Dos Santos Gomes, Isabel Cristina; Tinôco, Ricardo Salles; Quintela, Eliane Dias; Dos Reis Castilho, Leda; Freire, Denise Maria Guimarães
2018-03-01
The present study aimed to add value to palm oil by-products as substrates to efficiently produce conidia of Beauveria bassiana and Isaria javanica (Hypocreales: Cordycipitaceae) for biological control of the whitefly Bemisia tabaci (Hemiptera: Aleyrodidae), through a solid-state fermentation process using palm kernel cake and palm fiber as nutrient source and solid matrix, respectively. The optimum culture conditions yielded high concentrations of viable conidia after air-drying, when the fungi were grown on palm kernel cake (B. bassiana 7.65 × 10 9 and I. javanica 2.91 × 10 9 conidia g -1 dry substrate) after 6 days under optimal growth conditions set to 60% substrate moisture and 32 °C. Both fungal strains exhibited high efficacy against third-instar whitefly nymphs, inducing mortality up to 62.9 and 56.6% by B. bassiana and I. javanica, respectively, assessed after 9 days post-application in a screenhouse. Furthermore, we noted that insect mortality was strongly correlated with high atmospheric moisture, while B. bassiana appeared to require shorter accumulative hours under high moisture to kill whitefly nymphs compared to I. javanica. Our results underpin a feasible and cost-effective mass production method for aerial conidia, using palm kernel as the main substrate in order to produce efficacious fungal bioinsecticides against an invasive whitefly species in Brazil. Finally, our fermentation process may offer a sustainable and cost-effective means to produce eco-friendly mycoinsecticides, using an abundant agro-industrial by-product from Brazil that will ultimately assist in the integrated management of agricultural insect pests.
Parsimonious Wavelet Kernel Extreme Learning Machine
Wang Qin
2015-11-01
Full Text Available In this study, a parsimonious scheme for wavelet kernel extreme learning machine (named PWKELM was introduced by combining wavelet theory and a parsimonious algorithm into kernel extreme learning machine (KELM. In the wavelet analysis, bases that were localized in time and frequency to represent various signals effectively were used. Wavelet kernel extreme learning machine (WELM maximized its capability to capture the essential features in “frequency-rich” signals. The proposed parsimonious algorithm also incorporated significant wavelet kernel functions via iteration in virtue of Householder matrix, thus producing a sparse solution that eased the computational burden and improved numerical stability. The experimental results achieved from the synthetic dataset and a gas furnace instance demonstrated that the proposed PWKELM is efficient and feasible in terms of improving generalization accuracy and real time performance.
RKRD: Runtime Kernel Rootkit Detection
Grover, Satyajit; Khosravi, Hormuzd; Kolar, Divya; Moffat, Samuel; Kounavis, Michael E.
In this paper we address the problem of protecting computer systems against stealth malware. The problem is important because the number of known types of stealth malware increases exponentially. Existing approaches have some advantages for ensuring system integrity but sophisticated techniques utilized by stealthy malware can thwart them. We propose Runtime Kernel Rootkit Detection (RKRD), a hardware-based, event-driven, secure and inclusionary approach to kernel integrity that addresses some of the limitations of the state of the art. Our solution is based on the principles of using virtualization hardware for isolation, verifying signatures coming from trusted code as opposed to malware for scalability and performing system checks driven by events. Our RKRD implementation is guided by our goals of strong isolation, no modifications to target guest OS kernels, easy deployment, minimal infra-structure impact, and minimal performance overhead. We developed a system prototype and conducted a number of experiments which show that the per-formance impact of our solution is negligible.
Subspace Correction Methods for Total Variation and $\\ell_1$-Minimization
Fornasier, Massimo; Schö nlieb, Carola-Bibiane
2009-01-01
This paper is concerned with the numerical minimization of energy functionals in Hilbert spaces involving convex constraints coinciding with a seminorm for a subspace. The optimization is realized by alternating minimizations of the functional on a
A Realization of Temperature Monitoring System Based on Real-Time Kernel μC/OS and 1-wire Bus
Yanmei Qi
2013-06-01
Full Text Available The traditional temperature monitoring system generally adopt some analog sensors for collecting data and a microcontroller for processing data for the purpose of temperature monitoring. However, this back-fore ground system has the disadvantages that the system has poor real-time property and single function, the amount of sensors is not easy to expand, and the software system has a difficulty in upgrading. Aiming at these disadvantages, the system designed in this paper adopts brand-new hardware and software structures: a digitaltemperature sensor array is connected to 1-wire bus and communicated with a control core through 1-wire bus protocol, thus a great convenience is provided for the expansion of the sensor; a real-time operating system is introduced into the software, an application program capable of realizing various functions runs on the real-time kernel μC/OS-II platform. The application of the real-time kernel also provides a good lower layer interface for the late-stage software upgrading.
Raisali, G.R.
1992-01-01
A series of computer codes based on point kernel technique and also Monte Carlo method have been developed. These codes perform radiation transport calculations for irradiator systems having cartesian, cylindrical and mixed geometries. The monte Carlo calculations, the computer code 'EGS4' has been applied to a radiation processing type problem. This code has been acompanied by a specific user code. The set of codes developed include: GCELLS, DOSMAPM, DOSMAPC2 which simulate the radiation transport in gamma irradiator systems having cylinderical, cartesian, and mixed geometries, respectively. The program 'DOSMAP3' based on point kernel technique, has been also developed for dose rate mapping calculations in carrier type gamma irradiators. Another computer program 'CYLDETM' as a user code for EGS4 has been also developed to simulate dose variations near the interface of heterogeneous media in gamma irradiator systems. In addition a system of computer codes 'PRODMIX' has been developed which calculates the absorbed dose in the products with different densities. validation studies of the calculated results versus experimental dosimetry has been performed and good agreement has been obtained
On spectral subspaces and their applications to automorphism groups
Olesen, Dorte
1974-03-01
An attempt is made to give a survey of the theory of spectra and spectral subspaces of group representations in an abstract Banach space setting. The theory is applied to the groups of automorphisms of operator algebras (mostly C*-algebras) and some important results of interest for mathematical physicists are proved (restrictions of the bitransposed action, spectral subspaces for the transposed action on a C*-algebra, and positive states and representations of Rsup(n)) [fr
Filatov, Gleb; Bauwens, Bruno; Kertész-Farkas, Attila
2018-05-07
Bioinformatics studies often rely on similarity measures between sequence pairs, which often pose a bottleneck in large-scale sequence analysis. Here, we present a new convolutional kernel function for protein sequences called the LZW-Kernel. It is based on code words identified with the Lempel-Ziv-Welch (LZW) universal text compressor. The LZW-Kernel is an alignment-free method, it is always symmetric, is positive, always provides 1.0 for self-similarity and it can directly be used with Support Vector Machines (SVMs) in classification problems, contrary to normalized compression distance (NCD), which often violates the distance metric properties in practice and requires further techniques to be used with SVMs. The LZW-Kernel is a one-pass algorithm, which makes it particularly plausible for big data applications. Our experimental studies on remote protein homology detection and protein classification tasks reveal that the LZW-Kernel closely approaches the performance of the Local Alignment Kernel (LAK) and the SVM-pairwise method combined with Smith-Waterman (SW) scoring at a fraction of the time. Moreover, the LZW-Kernel outperforms the SVM-pairwise method when combined with BLAST scores, which indicates that the LZW code words might be a better basis for similarity measures than local alignment approximations found with BLAST. In addition, the LZW-Kernel outperforms n-gram based mismatch kernels, hidden Markov model based SAM and Fisher kernel, and protein family based PSI-BLAST, among others. Further advantages include the LZW-Kernel's reliance on a simple idea, its ease of implementation, and its high speed, three times faster than BLAST and several magnitudes faster than SW or LAK in our tests. LZW-Kernel is implemented as a standalone C code and is a free open-source program distributed under GPLv3 license and can be downloaded from https://github.com/kfattila/LZW-Kernel. akerteszfarkas@hse.ru. Supplementary data are available at Bioinformatics Online.
N. I. Didenko
2015-01-01
Full Text Available This paper presents a conceptual idea of the organization of management of development of the Arctic area of the Russian Federation in the form of a set of target subspace. Among the possible types of target subspace comprising the Arctic zone of the Russian Federation, allocated seven subspace: basic city mobile Camps, site production of mineral resources, recreational area, fishing area, the Northern Sea Route, infrastructure protection safe existence in the Arctic. The task of determining the most appropriate theoretical approach for the development of each target subspaces. To this end, the theoretical approaches of economic growth and development of the theory of "economic base» (Economic Base Theory; resource theory (Staple Theory; Theory sectors (Sector Theory; theory of growth poles (Growth Pole Theory; neoclassical theory (Neoclassical Growth Theory; theory of inter-regional trade (Interregional Trade Theory; theory of the commodity cycle; entrepreneurial theory (Entrepreneurship Theories.
Intelligent Control of a Sensor-Actuator System via Kernelized Least-Squares Policy Iteration
Bo Liu
2012-02-01
Full Text Available In this paper a new framework, called Compressive Kernelized Reinforcement Learning (CKRL, for computing near-optimal policies in sequential decision making with uncertainty is proposed via incorporating the non-adaptive data-independent Random Projections and nonparametric Kernelized Least-squares Policy Iteration (KLSPI. Random Projections are a fast, non-adaptive dimensionality reduction framework in which high-dimensionality data is projected onto a random lower-dimension subspace via spherically random rotation and coordination sampling. KLSPI introduce kernel trick into the LSPI framework for Reinforcement Learning, often achieving faster convergence and providing automatic feature selection via various kernel sparsification approaches. In this approach, policies are computed in a low-dimensional subspace generated by projecting the high-dimensional features onto a set of random basis. We first show how Random Projections constitute an efficient sparsification technique and how our method often converges faster than regular LSPI, while at lower computational costs. Theoretical foundation underlying this approach is a fast approximation of Singular Value Decomposition (SVD. Finally, simulation results are exhibited on benchmark MDP domains, which confirm gains both in computation time and in performance in large feature spaces.
Krylov subspace method for evaluating the self-energy matrices in electron transport calculations
Sørensen, Hans Henrik Brandenborg; Hansen, Per Christian; Petersen, D. E.
2008-01-01
We present a Krylov subspace method for evaluating the self-energy matrices used in the Green's function formulation of electron transport in nanoscale devices. A procedure based on the Arnoldi method is employed to obtain solutions of the quadratic eigenvalue problem associated with the infinite...... calculations. Numerical tests within a density functional theory framework are provided to validate the accuracy and robustness of the proposed method, which in most cases is an order of magnitude faster than conventional methods.......We present a Krylov subspace method for evaluating the self-energy matrices used in the Green's function formulation of electron transport in nanoscale devices. A procedure based on the Arnoldi method is employed to obtain solutions of the quadratic eigenvalue problem associated with the infinite...
Testa Antonia C
2010-10-01
Full Text Available Abstract Background Hitherto, risk prediction models for preoperative ultrasound-based diagnosis of ovarian tumors were dichotomous (benign versus malignant. We develop and validate polytomous models (models that predict more than two events to diagnose ovarian tumors as benign, borderline, primary invasive or metastatic invasive. The main focus is on how different types of models perform and compare. Methods A multi-center dataset containing 1066 women was used for model development and internal validation, whilst another multi-center dataset of 1938 women was used for temporal and external validation. Models were based on standard logistic regression and on penalized kernel-based algorithms (least squares support vector machines and kernel logistic regression. We used true polytomous models as well as combinations of dichotomous models based on the 'pairwise coupling' technique to produce polytomous risk estimates. Careful variable selection was performed, based largely on cross-validated c-index estimates. Model performance was assessed with the dichotomous c-index (i.e. the area under the ROC curve and a polytomous extension, and with calibration graphs. Results For all models, between 9 and 11 predictors were selected. Internal validation was successful with polytomous c-indexes between 0.64 and 0.69. For the best model dichotomous c-indexes were between 0.73 (primary invasive vs metastatic and 0.96 (borderline vs metastatic. On temporal and external validation, overall discrimination performance was good with polytomous c-indexes between 0.57 and 0.64. However, discrimination between primary and metastatic invasive tumors decreased to near random levels. Standard logistic regression performed well in comparison with advanced algorithms, and combining dichotomous models performed well in comparison with true polytomous models. The best model was a combination of dichotomous logistic regression models. This model is available online
Identification of Fusarium damaged wheat kernels using image analysis
Ondřej Jirsa
2011-01-01
Full Text Available Visual evaluation of kernels damaged by Fusarium spp. pathogens is labour intensive and due to a subjective approach, it can lead to inconsistencies. Digital imaging technology combined with appropriate statistical methods can provide much faster and more accurate evaluation of the visually scabby kernels proportion. The aim of the present study was to develop a discrimination model to identify wheat kernels infected by Fusarium spp. using digital image analysis and statistical methods. Winter wheat kernels from field experiments were evaluated visually as healthy or damaged. Deoxynivalenol (DON content was determined in individual kernels using an ELISA method. Images of individual kernels were produced using a digital camera on dark background. Colour and shape descriptors were obtained by image analysis from the area representing the kernel. Healthy and damaged kernels differed significantly in DON content and kernel weight. Various combinations of individual shape and colour descriptors were examined during the development of the model using linear discriminant analysis. In addition to basic descriptors of the RGB colour model (red, green, blue, very good classification was also obtained using hue from the HSL colour model (hue, saturation, luminance. The accuracy of classification using the developed discrimination model based on RGBH descriptors was 85 %. The shape descriptors themselves were not specific enough to distinguish individual kernels.
Implementing Kernel Methods Incrementally by Incremental Nonlinear Projection Trick.
Kwak, Nojun
2016-05-20
Recently, the nonlinear projection trick (NPT) was introduced enabling direct computation of coordinates of samples in a reproducing kernel Hilbert space. With NPT, any machine learning algorithm can be extended to a kernel version without relying on the so called kernel trick. However, NPT is inherently difficult to be implemented incrementally because an ever increasing kernel matrix should be treated as additional training samples are introduced. In this paper, an incremental version of the NPT (INPT) is proposed based on the observation that the centerization step in NPT is unnecessary. Because the proposed INPT does not change the coordinates of the old data, the coordinates obtained by INPT can directly be used in any incremental methods to implement a kernel version of the incremental methods. The effectiveness of the INPT is shown by applying it to implement incremental versions of kernel methods such as, kernel singular value decomposition, kernel principal component analysis, and kernel discriminant analysis which are utilized for problems of kernel matrix reconstruction, letter classification, and face image retrieval, respectively.
High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D
Wittmann, Roland
2017-01-25
We describe code optimization and parallelization procedures applied to the sequential overland flow solver FullSWOF2D. Major difficulties when simulating overland flows comprise dealing with high resolution datasets of large scale areas which either cannot be computed on a single node either due to limited amount of memory or due to too many (time step) iterations resulting from the CFL condition. We address these issues in terms of two major contributions. First, we demonstrate a generic step-by-step transformation of the second order finite volume scheme in FullSWOF2D towards MPI parallelization. Second, the computational kernels are optimized by the use of templates and a portable vectorization approach. We discuss the load imbalance of the flux computation due to dry and wet cells and propose a solution using an efficient cell counting approach. Finally, scalability results are shown for different test scenarios along with a flood simulation benchmark using the Shaheen II supercomputer.
Parameter optimization in the regularized kernel minimum noise fraction transformation
Nielsen, Allan Aasbjerg; Vestergaard, Jacob Schack
2012-01-01
Based on the original, linear minimum noise fraction (MNF) transformation and kernel principal component analysis, a kernel version of the MNF transformation was recently introduced. Inspired by we here give a simple method for finding optimal parameters in a regularized version of kernel MNF...... analysis. We consider the model signal-to-noise ratio (SNR) as a function of the kernel parameters and the regularization parameter. In 2-4 steps of increasingly refined grid searches we find the parameters that maximize the model SNR. An example based on data from the DLR 3K camera system is given....
Exploiting graph kernels for high performance biomedical relation extraction.
Panyam, Nagesh C; Verspoor, Karin; Cohn, Trevor; Ramamohanarao, Kotagiri
2018-01-30
Relation extraction from biomedical publications is an important task in the area of semantic mining of text. Kernel methods for supervised relation extraction are often preferred over manual feature engineering methods, when classifying highly ordered structures such as trees and graphs obtained from syntactic parsing of a sentence. Tree kernels such as the Subset Tree Kernel and Partial Tree Kernel have been shown to be effective for classifying constituency parse trees and basic dependency parse graphs of a sentence. Graph kernels such as the All Path Graph kernel (APG) and Approximate Subgraph Matching (ASM) kernel have been shown to be suitable for classifying general graphs with cycles, such as the enhanced dependency parse graph of a sentence. In this work, we present a high performance Chemical-Induced Disease (CID) relation extraction system. We present a comparative study of kernel methods for the CID task and also extend our study to the Protein-Protein Interaction (PPI) extraction task, an important biomedical relation extraction task. We discuss novel modifications to the ASM kernel to boost its performance and a method to apply graph kernels for extracting relations expressed in multiple sentences. Our system for CID relation extraction attains an F-score of 60%, without using external knowledge sources or task specific heuristic or rules. In comparison, the state of the art Chemical-Disease Relation Extraction system achieves an F-score of 56% using an ensemble of multiple machine learning methods, which is then boosted to 61% with a rule based system employing task specific post processing rules. For the CID task, graph kernels outperform tree kernels substantially, and the best performance is obtained with APG kernel that attains an F-score of 60%, followed by the ASM kernel at 57%. The performance difference between the ASM and APG kernels for CID sentence level relation extraction is not significant. In our evaluation of ASM for the PPI task, ASM
Adaptive metric kernel regression
Goutte, Cyril; Larsen, Jan
2000-01-01
Kernel smoothing is a widely used non-parametric pattern recognition technique. By nature, it suffers from the curse of dimensionality and is usually difficult to apply to high input dimensions. In this contribution, we propose an algorithm that adapts the input metric used in multivariate...... regression by minimising a cross-validation estimate of the generalisation error. This allows to automatically adjust the importance of different dimensions. The improvement in terms of modelling performance is illustrated on a variable selection task where the adaptive metric kernel clearly outperforms...
Adaptive Metric Kernel Regression
Goutte, Cyril; Larsen, Jan
1998-01-01
Kernel smoothing is a widely used nonparametric pattern recognition technique. By nature, it suffers from the curse of dimensionality and is usually difficult to apply to high input dimensions. In this paper, we propose an algorithm that adapts the input metric used in multivariate regression...... by minimising a cross-validation estimate of the generalisation error. This allows one to automatically adjust the importance of different dimensions. The improvement in terms of modelling performance is illustrated on a variable selection task where the adaptive metric kernel clearly outperforms the standard...
Wang, Shunfang; Nie, Bing; Yue, Kun; Fei, Yu; Li, Wenjia; Xu, Dongshu
2017-12-15
Kernel discriminant analysis (KDA) is a dimension reduction and classification algorithm based on nonlinear kernel trick, which can be novelly used to treat high-dimensional and complex biological data before undergoing classification processes such as protein subcellular localization. Kernel parameters make a great impact on the performance of the KDA model. Specifically, for KDA with the popular Gaussian kernel, to select the scale parameter is still a challenging problem. Thus, this paper introduces the KDA method and proposes a new method for Gaussian kernel parameter selection depending on the fact that the differences between reconstruction errors of edge normal samples and those of interior normal samples should be maximized for certain suitable kernel parameters. Experiments with various standard data sets of protein subcellular localization show that the overall accuracy of protein classification prediction with KDA is much higher than that without KDA. Meanwhile, the kernel parameter of KDA has a great impact on the efficiency, and the proposed method can produce an optimum parameter, which makes the new algorithm not only perform as effectively as the traditional ones, but also reduce the computational time and thus improve efficiency.
Detecting anomalies in crowded scenes via locality-constrained affine subspace coding
Fan, Yaxiang; Wen, Gongjian; Qiu, Shaohua; Li, Deren
2017-07-01
Video anomaly event detection is the process of finding an abnormal event deviation compared with the majority of normal or usual events. The main challenges are the high structure redundancy and the dynamic changes in the scenes that are in surveillance videos. To address these problems, we present a framework for anomaly detection and localization in videos that is based on locality-constrained affine subspace coding (LASC) and a model updating procedure. In our algorithm, LASC attempts to reconstruct the test sample by its top-k nearest subspaces, which are obtained by segmenting the normal samples space using a clustering method. A sample with a large reconstruction cost is detected as abnormal by setting a threshold. To adapt to the scene changes over time, a model updating strategy is proposed. We experiment on two public datasets: the UCSD dataset and the Avenue dataset. The results demonstrate that our method achieves competitive performance at a 700 fps on a single desktop PC.
Recursive Subspace Identification of AUV Dynamic Model under General Noise Assumption
Zheping Yan
2014-01-01
Full Text Available A recursive subspace identification algorithm for autonomous underwater vehicles (AUVs is proposed in this paper. Due to the advantages at handling nonlinearities and couplings, the AUV model investigated here is for the first time constructed as a Hammerstein model with nonlinear feedback in the linear part. To better take the environment and sensor noises into consideration, the identification problem is concerned as an errors-in-variables (EIV one which means that the identification procedure is under general noise assumption. In order to make the algorithm recursively, propagator method (PM based subspace approach is extended into EIV framework to form the recursive identification method called PM-EIV algorithm. With several identification experiments carried out by the AUV simulation platform, the proposed algorithm demonstrates its effectiveness and feasibility.
MODAL TRACKING of A Structural Device: A Subspace Identification Approach
Candy, J. V. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Franco, S. N. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Ruggiero, E. L. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Emmons, M. C. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Lopez, I. M. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Stoops, L. M. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
2017-03-20
Mechanical devices operating in an environment contaminated by noise, uncertainties, and extraneous disturbances lead to low signal-to-noise-ratios creating an extremely challenging processing problem. To detect/classify a device subsystem from noisy data, it is necessary to identify unique signatures or particular features. An obvious feature would be resonant (modal) frequencies emitted during its normal operation. In this report, we discuss a model-based approach to incorporate these physical features into a dynamic structure that can be used for such an identification. The approach we take after pre-processing the raw vibration data and removing any extraneous disturbances is to obtain a representation of the structurally unknown device along with its subsystems that capture these salient features. One approach is to recognize that unique modal frequencies (sinusoidal lines) appear in the estimated power spectrum that are solely characteristic of the device under investigation. Therefore, the objective of this effort is based on constructing a black box model of the device that captures these physical features that can be exploited to “diagnose” whether or not the particular device subsystem (track/detect/classify) is operating normally from noisy vibrational data. Here we discuss the application of a modern system identification approach based on stochastic subspace realization techniques capable of both (1) identifying the underlying black-box structure thereby enabling the extraction of structural modes that can be used for analysis and modal tracking as well as (2) indicators of condition and possible changes from normal operation.
Karri, Rama Rao; Sahu, J N
2018-01-15
Zn (II) is one the common pollutant among heavy metals found in industrial effluents. Removal of pollutant from industrial effluents can be accomplished by various techniques, out of which adsorption was found to be an efficient method. Applications of adsorption limits itself due to high cost of adsorbent. In this regard, a low cost adsorbent produced from palm oil kernel shell based agricultural waste is examined for its efficiency to remove Zn (II) from waste water and aqueous solution. The influence of independent process variables like initial concentration, pH, residence time, activated carbon (AC) dosage and process temperature on the removal of Zn (II) by palm kernel shell based AC from batch adsorption process are studied systematically. Based on the design of experimental matrix, 50 experimental runs are performed with each process variable in the experimental range. The optimal values of process variables to achieve maximum removal efficiency is studied using response surface methodology (RSM) and artificial neural network (ANN) approaches. A quadratic model, which consists of first order and second order degree regressive model is developed using the analysis of variance and RSM - CCD framework. The particle swarm optimization which is a meta-heuristic optimization is embedded on the ANN architecture to optimize the search space of neural network. The optimized trained neural network well depicts the testing data and validation data with R 2 equal to 0.9106 and 0.9279 respectively. The outcomes indicates that the superiority of ANN-PSO based model predictions over the quadratic model predictions provided by RSM. Copyright © 2017 Elsevier Ltd. All rights reserved.
Kernel methods for deep learning
Cho, Youngmin
2012-01-01
We introduce a new family of positive-definite kernels that mimic the computation in large neural networks. We derive the different members of this family by considering neural networks with different activation functions. Using these kernels as building blocks, we also show how to construct other positive-definite kernels by operations such as composition, multiplication, and averaging. We explore the use of these kernels in standard models of supervised learning, such as support vector mach...
Barndorff-Nielsen, Ole; Hansen, Peter Reinhard; Lunde, Asger
We propose a multivariate realised kernel to estimate the ex-post covariation of log-prices. We show this new consistent estimator is guaranteed to be positive semi-definite and is robust to measurement noise of certain types and can also handle non-synchronous trading. It is the first estimator...
Sommer, Stefan Horst; Lauze, Francois Bernard; Nielsen, Mads
2011-01-01
In the LDDMM framework, optimal warps for image registration are found as end-points of critical paths for an energy functional, and the EPDiff equations describe the evolution along such paths. The Large Deformation Diffeomorphic Kernel Bundle Mapping (LDDKBM) extension of LDDMM allows scale space...
Spafford, Eugene H.; Mckendry, Martin S.
1986-01-01
An overview of the internal structure of the Clouds kernel was presented. An indication of how these structures will interact in the prototype Clouds implementation is given. Many specific details have yet to be determined and await experimentation with an actual working system.
Embedded real-time operating system micro kernel design
Cheng, Xiao-hui; Li, Ming-qiang; Wang, Xin-zheng
2005-12-01
Embedded systems usually require a real-time character. Base on an 8051 microcontroller, an embedded real-time operating system micro kernel is proposed consisting of six parts, including a critical section process, task scheduling, interruption handle, semaphore and message mailbox communication, clock managent and memory managent. Distributed CPU and other resources are among tasks rationally according to the importance and urgency. The design proposed here provides the position, definition, function and principle of micro kernel. The kernel runs on the platform of an ATMEL AT89C51 microcontroller. Simulation results prove that the designed micro kernel is stable and reliable and has quick response while operating in an application system.
Reverse time migration by Krylov subspace reduced order modeling
Basir, Hadi Mahdavi; Javaherian, Abdolrahim; Shomali, Zaher Hossein; Firouz-Abadi, Roohollah Dehghani; Gholamy, Shaban Ali
2018-04-01
Imaging is a key step in seismic data processing. To date, a myriad of advanced pre-stack depth migration approaches have been developed; however, reverse time migration (RTM) is still considered as the high-end imaging algorithm. The main limitations associated with the performance cost of reverse time migration are the intensive computation of the forward and backward simulations, time consumption, and memory allocation related to imaging condition. Based on the reduced order modeling, we proposed an algorithm, which can be adapted to all the aforementioned factors. Our proposed method benefit from Krylov subspaces method to compute certain mode shapes of the velocity model computed by as an orthogonal base of reduced order modeling. Reverse time migration by reduced order modeling is helpful concerning the highly parallel computation and strongly reduces the memory requirement of reverse time migration. The synthetic model results showed that suggested method can decrease the computational costs of reverse time migration by several orders of magnitudes, compared with reverse time migration by finite element method.
Supervised orthogonal discriminant subspace projects learning for face recognition.
Chen, Yu; Xu, Xiao-Hong
2014-02-01
In this paper, a new linear dimension reduction method called supervised orthogonal discriminant subspace projection (SODSP) is proposed, which addresses high-dimensionality of data and the small sample size problem. More specifically, given a set of data points in the ambient space, a novel weight matrix that describes the relationship between the data points is first built. And in order to model the manifold structure, the class information is incorporated into the weight matrix. Based on the novel weight matrix, the local scatter matrix as well as non-local scatter matrix is defined such that the neighborhood structure can be preserved. In order to enhance the recognition ability, we impose an orthogonal constraint into a graph-based maximum margin analysis, seeking to find a projection that maximizes the difference, rather than the ratio between the non-local scatter and the local scatter. In this way, SODSP naturally avoids the singularity problem. Further, we develop an efficient and stable algorithm for implementing SODSP, especially, on high-dimensional data set. Moreover, the theoretical analysis shows that LPP is a special instance of SODSP by imposing some constraints. Experiments on the ORL, Yale, Extended Yale face database B and FERET face database are performed to test and evaluate the proposed algorithm. The results demonstrate the effectiveness of SODSP. Copyright © 2013 Elsevier Ltd. All rights reserved.
A Variational Approach to Video Registration with Subspace Constraints.
Garg, Ravi; Roussos, Anastasios; Agapito, Lourdes
2013-01-01
This paper addresses the problem of non-rigid video registration, or the computation of optical flow from a reference frame to each of the subsequent images in a sequence, when the camera views deformable objects. We exploit the high correlation between 2D trajectories of different points on the same non-rigid surface by assuming that the displacement of any point throughout the sequence can be expressed in a compact way as a linear combination of a low-rank motion basis. This subspace constraint effectively acts as a trajectory regularization term leading to temporally consistent optical flow. We formulate it as a robust soft constraint within a variational framework by penalizing flow fields that lie outside the low-rank manifold. The resulting energy functional can be decoupled into the optimization of the brightness constancy and spatial regularization terms, leading to an efficient optimization scheme. Additionally, we propose a novel optimization scheme for the case of vector valued images, based on the dualization of the data term. This allows us to extend our approach to deal with colour images which results in significant improvements on the registration results. Finally, we provide a new benchmark dataset, based on motion capture data of a flag waving in the wind, with dense ground truth optical flow for evaluation of multi-frame optical flow algorithms for non-rigid surfaces. Our experiments show that our proposed approach outperforms state of the art optical flow and dense non-rigid registration algorithms.
Chen, Xing; Zeng, Yan; Yin, Yuehong
2017-06-01
Transparent control is still highly challenging for robotic exoskeletons, especially when a simple strategy is expected for a large-impedance device. To improve the transparency for late-phase rehabilitation when "patient-in-charge" mode is necessary, this paper aims at adaptive identification of human motor intent, and proposed an iterative prediction-compensation motion control scheme for an exoskeleton knee joint. Based on the analysis of human-machine interactive mechanism (HMIM) and the semiphenomenological biomechanical model of muscle, an online adaptive predicting controller is designed using a focused time-delay neural network (FTDNN) with the inputs of electromyography (EMG), position and interactive force, where the activation level of muscle is estimated from EMG using a novel energy kernel method. The compensating controller is designed using the normative force-position control paradigm. Initial experiments on the human-machine integrated knee system validated the effectiveness and ease of use of the proposed control scheme.
Application of Earthquake Subspace Detectors at Kilauea and Mauna Loa Volcanoes, Hawai`i
Okubo, P.; Benz, H.; Yeck, W.
2016-12-01
Recent studies have demonstrated the capabilities of earthquake subspace detectors for detailed cataloging and tracking of seismicity in a number of regions and settings. We are exploring the application of subspace detectors at the United States Geological Survey's Hawaiian Volcano Observatory (HVO) to analyze seismicity at Kilauea and Mauna Loa volcanoes. Elevated levels of microseismicity and occasional swarms of earthquakes associated with active volcanism here present cataloging challenges due the sheer numbers of earthquakes and an intrinsically low signal-to-noise environment featuring oceanic microseism and volcanic tremor in the ambient seismic background. With high-quality continuous recording of seismic data at HVO, we apply subspace detectors (Harris and Dodge, 2011, Bull. Seismol. Soc. Am., doi: 10.1785/0120100103) during intervals of noteworthy seismicity. Waveform templates are drawn from Magnitude 2 and larger earthquakes within clusters of earthquakes cataloged in the HVO seismic database. At Kilauea, we focus on seismic swarms in the summit caldera region where, despite continuing eruptions from vents in the summit region and in the east rift zone, geodetic measurements reflect a relatively inflated volcanic state. We also focus on seismicity beneath and adjacent to Mauna Loa's summit caldera that appears to be associated with geodetic expressions of gradual volcanic inflation, and where precursory seismicity clustered prior to both Mauna Loa's most recent eruptions in 1975 and 1984. We recover several times more earthquakes with the subspace detectors - down to roughly 2 magnitude units below the templates, based on relative amplitudes - compared to the numbers of cataloged earthquakes. The increased numbers of detected earthquakes in these clusters, and the ability to associate and locate them, allow us to infer details of the spatial and temporal distributions and possible variations in stresses within these key regions of the volcanoes.
Honório, Leonardo M; Barbosa, Daniele A; Oliveira, Edimar J; Garcia, Paulo A Nepomuceno; Santos, Murillo F
2018-03-01
This work presents a new approach for solving classification and learning problems. The Successive Geometric Segmentation technique is applied to encapsulate large datasets by using a series of Oriented Bounding Hyper Box (OBHBs). Each OBHB is obtained through linear separation analysis and each one represents a specific region in a pattern's solution space. Also, each OBHB can be seen as a data abstraction layer and be considered as an individual Kernel. Thus, it is possible by applying a quadratic discriminant function, to assemble a set of nonlinear surfaces separating each desirable pattern. This approach allows working with large datasets using high speed linear analysis tools and yet providing a very accurate non-linear classifier as final result. The methodology was tested using the UCI Machine Learning repository and a Power Transformer Fault Diagnosis real scenario problem. The results were compared with different approaches provided by literature and, finally, the potential and further applications of the methodology were also discussed. Copyright © 2018 ISA. Published by Elsevier Ltd. All rights reserved.
Analog forecasting with dynamics-adapted kernels
Zhao, Zhizhen; Giannakis, Dimitrios
2016-09-01
Analog forecasting is a nonparametric technique introduced by Lorenz in 1969 which predicts the evolution of states of a dynamical system (or observables defined on the states) by following the evolution of the sample in a historical record of observations which most closely resembles the current initial data. Here, we introduce a suite of forecasting methods which improve traditional analog forecasting by combining ideas from kernel methods developed in harmonic analysis and machine learning and state-space reconstruction for dynamical systems. A key ingredient of our approach is to replace single-analog forecasting with weighted ensembles of analogs constructed using local similarity kernels. The kernels used here employ a number of dynamics-dependent features designed to improve forecast skill, including Takens’ delay-coordinate maps (to recover information in the initial data lost through partial observations) and a directional dependence on the dynamical vector field generating the data. Mathematically, our approach is closely related to kernel methods for out-of-sample extension of functions, and we discuss alternative strategies based on the Nyström method and the multiscale Laplacian pyramids technique. We illustrate these techniques in applications to forecasting in a low-order deterministic model for atmospheric dynamics with chaotic metastability, and interannual-scale forecasting in the North Pacific sector of a comprehensive climate model. We find that forecasts based on kernel-weighted ensembles have significantly higher skill than the conventional approach following a single analog.
de Almeida, Valber Elias; de Araújo Gomes, Adriano; de Sousa Fernandes, David Douglas; Goicoechea, Héctor Casimiro; Galvão, Roberto Kawakami Harrop; Araújo, Mario Cesar Ugulino
2018-05-01
This paper proposes a new variable selection method for nonlinear multivariate calibration, combining the Successive Projections Algorithm for interval selection (iSPA) with the Kernel Partial Least Squares (Kernel-PLS) modelling technique. The proposed iSPA-Kernel-PLS algorithm is employed in a case study involving a Vis-NIR spectrometric dataset with complex nonlinear features. The analytical problem consists of determining Brix and sucrose content in samples from a sugar production system, on the basis of transflectance spectra. As compared to full-spectrum Kernel-PLS, the iSPA-Kernel-PLS models involve a smaller number of variables and display statistically significant superiority in terms of accuracy and/or bias in the predictions. Published by Elsevier B.V.
Viscosity kernel of molecular fluids
Puscasu, Ruslan; Todd, Billy; Daivis, Peter
2010-01-01
, temperature, and chain length dependencies of the reciprocal and real-space viscosity kernels are presented. We find that the density has a major effect on the shape of the kernel. The temperature range and chain lengths considered here have by contrast less impact on the overall normalized shape. Functional...... forms that fit the wave-vector-dependent kernel data over a large density and wave-vector range have also been tested. Finally, a structural normalization of the kernels in physical space is considered. Overall, the real-space viscosity kernel has a width of roughly 3–6 atomic diameters, which means...
Subspace identification of Hammer stein models using support vector machines
Al-Dhaifallah, Mujahed
2011-01-01
System identification is the art of finding mathematical tools and algorithms that build an appropriate mathematical model of a system from measured input and output data. Hammerstein model, consisting of a memoryless nonlinearity followed by a dynamic linear element, is often a good trade-off as it can represent some dynamic nonlinear systems very accurately, but is nonetheless quite simple. Moreover, the extensive knowledge about LTI system representations can be applied to the dynamic linear block. On the other hand, finding an effective representation for the nonlinearity is an active area of research. Recently, support vector machines (SVMs) and least squares support vector machines (LS-SVMs) have demonstrated powerful abilities in approximating linear and nonlinear functions. In contrast with other approximation methods, SVMs do not require a-priori structural information. Furthermore, there are well established methods with guaranteed convergence (ordinary least squares, quadratic programming) for fitting LS-SVMs and SVMs. The general objective of this research is to develop new subspace algorithms for Hammerstein systems based on SVM regression.
Hadamard Kernel SVM with applications for breast cancer outcome predictions.
Jiang, Hao; Ching, Wai-Ki; Cheung, Wai-Shun; Hou, Wenpin; Yin, Hong
2017-12-21
Breast cancer is one of the leading causes of deaths for women. It is of great necessity to develop effective methods for breast cancer detection and diagnosis. Recent studies have focused on gene-based signatures for outcome predictions. Kernel SVM for its discriminative power in dealing with small sample pattern recognition problems has attracted a lot attention. But how to select or construct an appropriate kernel for a specified problem still needs further investigation. Here we propose a novel kernel (Hadamard Kernel) in conjunction with Support Vector Machines (SVMs) to address the problem of breast cancer outcome prediction using gene expression data. Hadamard Kernel outperform the classical kernels and correlation kernel in terms of Area under the ROC Curve (AUC) values where a number of real-world data sets are adopted to test the performance of different methods. Hadamard Kernel SVM is effective for breast cancer predictions, either in terms of prognosis or diagnosis. It may benefit patients by guiding therapeutic options. Apart from that, it would be a valuable addition to the current SVM kernel families. We hope it will contribute to the wider biology and related communities.
Independence and totalness of subspaces in phase space methods
Vourdas, A.
2018-04-01
The concepts of independence and totalness of subspaces are introduced in the context of quasi-probability distributions in phase space, for quantum systems with finite-dimensional Hilbert space. It is shown that due to the non-distributivity of the lattice of subspaces, there are various levels of independence, from pairwise independence up to (full) independence. Pairwise totalness, totalness and other intermediate concepts are also introduced, which roughly express that the subspaces overlap strongly among themselves, and they cover the full Hilbert space. A duality between independence and totalness, that involves orthocomplementation (logical NOT operation), is discussed. Another approach to independence is also studied, using Rota's formalism on independent partitions of the Hilbert space. This is used to define informational independence, which is proved to be equivalent to independence. As an application, the pentagram (used in discussions on contextuality) is analysed using these concepts.
Reduced-Rank Adaptive Filtering Using Krylov Subspace
Sergueï Burykh
2003-01-01
Full Text Available A unified view of several recently introduced reduced-rank adaptive filters is presented. As all considered methods use Krylov subspace for rank reduction, the approach taken in this work is inspired from Krylov subspace methods for iterative solutions of linear systems. The alternative interpretation so obtained is used to study the properties of each considered technique and to relate one reduced-rank method to another as well as to algorithms used in computational linear algebra. Practical issues are discussed and low-complexity versions are also included in our study. It is believed that the insight developed in this paper can be further used to improve existing reduced-rank methods according to known results in the domain of Krylov subspace methods.
A generalized Schwinger boson mapping with a physical subspace
Scholtz, F.G.; Geyer, H.B.
1988-01-01
We investigate the existence of a physical subspace for generalized Schwinger boson mappings of SO(2n+1) contains SO(2n) in view of previous observations by Marshalek and the recent construction of such a mapping and subspace for SO(8) by Kaup. It is shown that Kaup's construction can be attributed to the existence of a unique SO(8) automorphism. We proceed to construct a generalized Schwinger-type mapping for SO(2n+1) contains SO(2n) which, in contrast to a similar attempt by Yamamura and Nishiyama, indeed has a corresponding physical subspace. This new mapping includes in the special case of SO(8) the mapping by Kaup which is equivalent to the one given by Yamamura and Nishiyama for n=4. Nevertheless, we indicate the limitations of the generalized Schwinger mapping regarding its applicability to situations where one seeks to establish a direct link between phenomenological boson models and an underlying fermion microscopy. (orig.)
Emil Racolța
2014-11-01
Full Text Available The known confectionery spreadable cream product category includes well-known cocoa - hazelnut pastes as well as peanut butter, products that became very popular in the last decades due to their pleasant taste and ease of eating. However, health constraints appeared both hazelnut and peanut are food allergens, while cocoa excites central nervous system (CNS and on everyday consumption causes dependence. The aim of this work was to characterize the aroma and sensory of an innovative product that belongs to the same confectionery spreadable cream product category. Six spreadable cream prototypes were produced by using sugar, roasted sunflower kernel, carob or cocoa powder, palm or coconut fat and, lecithin. The obtained samples were firstly analyzed by using the nine point hedonic scale test. The volatile compounds profile analysis (“In Tube Extraction”- GC-MS was performed on the best samples (in terms of sensory containing cocoa or carob powder, as well as a control. The main volatile compound of all three samples was pinene (42-51% which is a characteristic flavor of turpentine, wood. Acetophenone instead (20-25% gives flavors of almond, floral, sweetish. Benzaldehyde (8.11-9.73% is characteristic for almond flavor with hints of caramel. The study revealed that the analyzed spreadable creams have similar volatile profiles, even if carob and cocoa powder showed different volatile compounds profiles, with the major compound for both being Propanoic acid, 2-methyl. Thus, with similar taste to cocoa sample, carob-sunflower spreadable cream is an alternative that not include ingredients with allergic potential or CNS stimulants.
Effect of Palm Kernel Cake Replacement and Enzyme ...
A feeding trial which lasted for twelve weeks was conducted to study the performance of finisher pigs fed five different levels of palm kernel cake replacement for maize (0%, 40%, 40%, 60%, 60%) in a maize-palm kernel cake based ration with or without enzyme supplementation. It was a completely randomized design ...
Variable Kernel Density Estimation
Terrell, George R.; Scott, David W.
1992-01-01
We investigate some of the possibilities for improvement of univariate and multivariate kernel density estimates by varying the window over the domain of estimation, pointwise and globally. Two general approaches are to vary the window width by the point of estimation and by point of the sample observation. The first possibility is shown to be of little efficacy in one variable. In particular, nearest-neighbor estimators in all versions perform poorly in one and two dimensions, but begin to b...
Steerability of Hermite Kernel
Yang, Bo; Flusser, Jan; Suk, Tomáš
2013-01-01
Roč. 27, č. 4 (2013), 1354006-1-1354006-25 ISSN 0218-0014 R&D Projects: GA ČR GAP103/11/1552 Institutional support: RVO:67985556 Keywords : Hermite polynomials * Hermite kernel * steerability * adaptive filtering Subject RIV: JD - Computer Applications, Robotics Impact factor: 0.558, year: 2013 http://library.utia.cas.cz/separaty/2013/ZOI/yang-0394387. pdf
Kernel learning at the first level of inference.
Cawley, Gavin C; Talbot, Nicola L C
2014-05-01
Kernel learning methods, whether Bayesian or frequentist, typically involve multiple levels of inference, with the coefficients of the kernel expansion being determined at the first level and the kernel and regularisation parameters carefully tuned at the second level, a process known as model selection. Model selection for kernel machines is commonly performed via optimisation of a suitable model selection criterion, often based on cross-validation or theoretical performance bounds. However, if there are a large number of kernel parameters, as for instance in the case of automatic relevance determination (ARD), there is a substantial risk of over-fitting the model selection criterion, resulting in poor generalisation performance. In this paper we investigate the possibility of learning the kernel, for the Least-Squares Support Vector Machine (LS-SVM) classifier, at the first level of inference, i.e. parameter optimisation. The kernel parameters and the coefficients of the kernel expansion are jointly optimised at the first level of inference, minimising a training criterion with an additional regularisation term acting on the kernel parameters. The key advantage of this approach is that the values of only two regularisation parameters need be determined in model selection, substantially alleviating the problem of over-fitting the model selection criterion. The benefits of this approach are demonstrated using a suite of synthetic and real-world binary classification benchmark problems, where kernel learning at the first level of inference is shown to be statistically superior to the conventional approach, improves on our previous work (Cawley and Talbot, 2007) and is competitive with Multiple Kernel Learning approaches, but with reduced computational expense. Copyright © 2014 Elsevier Ltd. All rights reserved.
Fast regularizing sequential subspace optimization in Banach spaces
Schöpfer, F; Schuster, T
2009-01-01
We are concerned with fast computations of regularized solutions of linear operator equations in Banach spaces in case only noisy data are available. To this end we modify recently developed sequential subspace optimization methods in such a way that the therein employed Bregman projections onto hyperplanes are replaced by Bregman projections onto stripes whose width is in the order of the noise level
Recursive subspace identification for in flight modal analysis of airplanes
De Cock , Katrien; Mercère , Guillaume; De Moor , Bart
2006-01-01
International audience; In this paper recursive subspace identification algorithms are applied to track the modal parameters of airplanes on-line during test flights. The ability to track changes in the damping ratios and the influence of the forgetting factor are studied through simulations.
Von Neumann algebras as complemented subspaces of B(H)
Christensen, Erik; Wang, Liguang
2014-01-01
Let M be a von Neumann algebra of type II1 which is also a complemented subspace of B( H). We establish an algebraic criterion, which ensures that M is an injective von Neumann algebra. As a corollary we show that if M is a complemented factor of type II1 on a Hilbert space H, then M is injective...
Intrinsic Grassmann Averages for Online Linear and Robust Subspace Learning
Chakraborty, Rudrasis; Hauberg, Søren; Vemuri, Baba C.
2017-01-01
Principal Component Analysis (PCA) is a fundamental method for estimating a linear subspace approximation to high-dimensional data. Many algorithms exist in literature to achieve a statistically robust version of PCA called RPCA. In this paper, we present a geometric framework for computing the p...
Active Subspace Methods for Data-Intensive Inverse Problems
Wang, Qiqi [Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
2017-04-27
The project has developed theory and computational tools to exploit active subspaces to reduce the dimension in statistical calibration problems. This dimension reduction enables MCMC methods to calibrate otherwise intractable models. The same theoretical and computational tools can also reduce the measurement dimension for calibration problems that use large stores of data.
Subspace identification of distributed clusters of homogeneous systems
Yu, C.; Verhaegen, M.H.G.
2017-01-01
This note studies the identification of a network comprised of interconnected clusters of LTI systems. Each cluster consists of homogeneous dynamical systems, and its interconnections with the rest of the network are unmeasurable. A subspace identification method is proposed for identifying a single
Symbol recognition with kernel density matching.
Zhang, Wan; Wenyin, Liu; Zhang, Kun
2006-12-01
We propose a novel approach to similarity assessment for graphic symbols. Symbols are represented as 2D kernel densities and their similarity is measured by the Kullback-Leibler divergence. Symbol orientation is found by gradient-based angle searching or independent component analysis. Experimental results show the outstanding performance of this approach in various situations.
Reproducing kernel Hilbert spaces of Gaussian priors
Vaart, van der A.W.; Zanten, van J.H.; Clarke, B.; Ghosal, S.
2008-01-01
We review definitions and properties of reproducing kernel Hilbert spaces attached to Gaussian variables and processes, with a view to applications in nonparametric Bayesian statistics using Gaussian priors. The rate of contraction of posterior distributions based on Gaussian priors can be described
Smolka, Gert
1994-01-01
Oz is a concurrent language providing for functional, object-oriented, and constraint programming. This paper defines Kernel Oz, a semantically complete sublanguage of Oz. It was an important design requirement that Oz be definable by reduction to a lean kernel language. The definition of Kernel Oz introduces three essential abstractions: the Oz universe, the Oz calculus, and the actor model. The Oz universe is a first-order structure defining the values and constraints Oz computes with. The ...
2010-01-01
... 7 Agriculture 8 2010-01-01 2010-01-01 false Edible kernel. 981.7 Section 981.7 Agriculture... Regulating Handling Definitions § 981.7 Edible kernel. Edible kernel means a kernel, piece, or particle of almond kernel that is not inedible. [41 FR 26852, June 30, 1976] ...
7 CFR 981.408 - Inedible kernel.
2010-01-01
... 7 Agriculture 8 2010-01-01 2010-01-01 false Inedible kernel. 981.408 Section 981.408 Agriculture... Administrative Rules and Regulations § 981.408 Inedible kernel. Pursuant to § 981.8, the definition of inedible kernel is modified to mean a kernel, piece, or particle of almond kernel with any defect scored as...
7 CFR 981.8 - Inedible kernel.
2010-01-01
... 7 Agriculture 8 2010-01-01 2010-01-01 false Inedible kernel. 981.8 Section 981.8 Agriculture... Regulating Handling Definitions § 981.8 Inedible kernel. Inedible kernel means a kernel, piece, or particle of almond kernel with any defect scored as serious damage, or damage due to mold, gum, shrivel, or...
3-D waveform tomography sensitivity kernels for anisotropic media
Djebbi, Ramzi
2014-01-01
The complications in anisotropic multi-parameter inversion lie in the trade-off between the different anisotropy parameters. We compute the tomographic waveform sensitivity kernels for a VTI acoustic medium perturbation as a tool to investigate this ambiguity between the different parameters. We use dynamic ray tracing to efficiently handle the expensive computational cost for 3-D anisotropic models. Ray tracing provides also the ray direction information necessary for conditioning the sensitivity kernels to handle anisotropy. The NMO velocity and η parameter kernels showed a maximum sensitivity for diving waves which results in a relevant choice of those parameters in wave equation tomography. The δ parameter kernel showed zero sensitivity; therefore it can serve as a secondary parameter to fit the amplitude in the acoustic anisotropic inversion. Considering the limited penetration depth of diving waves, migration velocity analysis based kernels are introduced to fix the depth ambiguity with reflections and compute sensitivity maps in the deeper parts of the model.
Barndorff-Nielsen, Ole Eiler; Hansen, Peter Reinhard; Lunde, Asger
2011-01-01
We propose a multivariate realised kernel to estimate the ex-post covariation of log-prices. We show this new consistent estimator is guaranteed to be positive semi-definite and is robust to measurement error of certain types and can also handle non-synchronous trading. It is the first estimator...... which has these three properties which are all essential for empirical work in this area. We derive the large sample asymptotics of this estimator and assess its accuracy using a Monte Carlo study. We implement the estimator on some US equity data, comparing our results to previous work which has used...
Sahutaya Pongprayoon
2010-12-01
Full Text Available To investigate the effects of inclusion of palm kernel cake (PKC in the diets on intake, digestibility, rumen fermentationcharacteristics, nitrogen balance and microbial N supply, five goats (initial BW = 20±1 kg were randomly assigned to a55 Latin square design to receive five diets, T1 = concentrate with 15% PKC, T2 = 25% PKC, T3 = 35% PKC, T4 = 45% PKCand T5 = 55% PKC, of dietary dry matter, respectively. Plicatulum hay was offered ad libitum as the roughage. A metabolismtrial lasted for 21 days during which live weight changes and feed intakes were measured. Based on this experiment, therewere no significant differences (p>0.05 among treatment groups regarding dry matter (DM intake and digestion coefficientsof DM, organic matter, crude protein, neutral detergent fiber and acid detergent fiber, except in T4 and T5 (45 and 55% PKCwhich had lower (p0.05, however the concentration of total volatile fatty acids and protozoal populations were slightly lower forgoats fed inclusion of 45-55% PKC as compared with other treatments. Based on this experiment, it could be concluded thatthe optimal level of PKC in concentrate should be 15-35% for goats fed with plicatulum hay and that it may be an effectivemeans of exploiting the use of local feed resources for goat production.
Kernel Temporal Differences for Neural Decoding
Bae, Jihye; Sanchez Giraldo, Luis G.; Pohlmeyer, Eric A.; Francis, Joseph T.; Sanchez, Justin C.; Príncipe, José C.
2015-01-01
We study the feasibility and capability of the kernel temporal difference (KTD)(λ) algorithm for neural decoding. KTD(λ) is an online, kernel-based learning algorithm, which has been introduced to estimate value functions in reinforcement learning. This algorithm combines kernel-based representations with the temporal difference approach to learning. One of our key observations is that by using strictly positive definite kernels, algorithm's convergence can be guaranteed for policy evaluation. The algorithm's nonlinear functional approximation capabilities are shown in both simulations of policy evaluation and neural decoding problems (policy improvement). KTD can handle high-dimensional neural states containing spatial-temporal information at a reasonable computational complexity allowing real-time applications. When the algorithm seeks a proper mapping between a monkey's neural states and desired positions of a computer cursor or a robot arm, in both open-loop and closed-loop experiments, it can effectively learn the neural state to action mapping. Finally, a visualization of the coadaptation process between the decoder and the subject shows the algorithm's capabilities in reinforcement learning brain machine interfaces. PMID:25866504
Localized Multiple Kernel Learning Via Sample-Wise Alternating Optimization.
Han, Yina; Yang, Kunde; Ma, Yuanliang; Liu, Guizhong
2014-01-01
Our objective is to train support vector machines (SVM)-based localized multiple kernel learning (LMKL), using the alternating optimization between the standard SVM solvers with the local combination of base kernels and the sample-specific kernel weights. The advantage of alternating optimization developed from the state-of-the-art MKL is the SVM-tied overall complexity and the simultaneous optimization on both the kernel weights and the classifier. Unfortunately, in LMKL, the sample-specific character makes the updating of kernel weights a difficult quadratic nonconvex problem. In this paper, starting from a new primal-dual equivalence, the canonical objective on which state-of-the-art methods are based is first decomposed into an ensemble of objectives corresponding to each sample, namely, sample-wise objectives. Then, the associated sample-wise alternating optimization method is conducted, in which the localized kernel weights can be independently obtained by solving their exclusive sample-wise objectives, either linear programming (for l1-norm) or with closed-form solutions (for lp-norm). At test time, the learnt kernel weights for the training data are deployed based on the nearest-neighbor rule. Hence, to guarantee their generality among the test part, we introduce the neighborhood information and incorporate it into the empirical loss when deriving the sample-wise objectives. Extensive experiments on four benchmark machine learning datasets and two real-world computer vision datasets demonstrate the effectiveness and efficiency of the proposed algorithm.
Yang Zhang
2015-11-01
Full Text Available Prognostics is necessary to ensure the reliability and safety of lithium-ion batteries for hybrid electric vehicles or satellites. This process can be achieved by capacity estimation, which is a direct fading indicator for assessing the state of health of a battery. However, the capacity of a lithium-ion battery onboard is difficult to monitor. This paper presents a data-driven approach for online capacity estimation. First, six novel features are extracted from cyclic charge/discharge cycles and used as indirect health indicators. An adaptive multi-kernel relevance machine (MKRVM based on accelerated particle swarm optimization algorithm is used to determine the optimal parameters of MKRVM and characterize the relationship between extracted features and battery capacity. The overall estimation process comprises offline and online stages. A supervised learning step in the offline stage is established for model verification to ensure the generalizability of MKRVM for online application. Cross-validation is further conducted to validate the performance of the proposed model. Experiment and comparison results show the effectiveness, accuracy, efficiency, and robustness of the proposed approach for online capacity estimation of lithium-ion batteries.
Silitonga, Arridina Susan; Hassan, Masjuki Haji; Ong, Hwai Chyuan; Kusumo, Fitranto
2017-11-01
The purpose of this study is to investigate the performance, emission and combustion characteristics of a four-cylinder common-rail turbocharged diesel engine fuelled with Jatropha curcas biodiesel-diesel blends. A kernel-based extreme learning machine (KELM) model is developed in this study using MATLAB software in order to predict the performance, combustion and emission characteristics of the engine. To acquire the data for training and testing the KELM model, the engine speed was selected as the input parameter, whereas the performance, exhaust emissions and combustion characteristics were chosen as the output parameters of the KELM model. The performance, emissions and combustion characteristics predicted by the KELM model were validated by comparing the predicted data with the experimental data. The results show that the coefficient of determination of the parameters is within a range of 0.9805-0.9991 for both the KELM model and the experimental data. The mean absolute percentage error is within a range of 0.1259-2.3838. This study shows that KELM modelling is a useful technique in biodiesel production since it facilitates scientists and researchers to predict the performance, exhaust emissions and combustion characteristics of internal combustion engines with high accuracy.
A General Algorithm for Reusing Krylov Subspace Information. I. Unsteady Navier-Stokes
Carpenter, Mark H.; Vuik, C.; Lucas, Peter; vanGijzen, Martin; Bijl, Hester
2010-01-01
A general algorithm is developed that reuses available information to accelerate the iterative convergence of linear systems with multiple right-hand sides A x = b (sup i), which are commonly encountered in steady or unsteady simulations of nonlinear equations. The algorithm is based on the classical GMRES algorithm with eigenvector enrichment but also includes a Galerkin projection preprocessing step and several novel Krylov subspace reuse strategies. The new approach is applied to a set of test problems, including an unsteady turbulent airfoil, and is shown in some cases to provide significant improvement in computational efficiency relative to baseline approaches.
Global Polynomial Kernel Hazard Estimation
Hiabu, Munir; Miranda, Maria Dolores Martínez; Nielsen, Jens Perch
2015-01-01
This paper introduces a new bias reducing method for kernel hazard estimation. The method is called global polynomial adjustment (GPA). It is a global correction which is applicable to any kernel hazard estimator. The estimator works well from a theoretical point of view as it asymptotically redu...
Bayesian Kernel Mixtures for Counts.
Canale, Antonio; Dunson, David B
2011-12-01
Although Bayesian nonparametric mixture models for continuous data are well developed, there is a limited literature on related approaches for count data. A common strategy is to use a mixture of Poissons, which unfortunately is quite restrictive in not accounting for distributions having variance less than the mean. Other approaches include mixing multinomials, which requires finite support, and using a Dirichlet process prior with a Poisson base measure, which does not allow smooth deviations from the Poisson. As a broad class of alternative models, we propose to use nonparametric mixtures of rounded continuous kernels. An efficient Gibbs sampler is developed for posterior computation, and a simulation study is performed to assess performance. Focusing on the rounded Gaussian case, we generalize the modeling framework to account for multivariate count data, joint modeling with continuous and categorical variables, and other complications. The methods are illustrated through applications to a developmental toxicity study and marketing data. This article has supplementary material online.
Explicit signal to noise ratio in reproducing kernel Hilbert spaces
Gomez-Chova, Luis; Nielsen, Allan Aasbjerg; Camps-Valls, Gustavo
2011-01-01
This paper introduces a nonlinear feature extraction method based on kernels for remote sensing data analysis. The proposed approach is based on the minimum noise fraction (MNF) transform, which maximizes the signal variance while also minimizing the estimated noise variance. We here propose...... an alternative kernel MNF (KMNF) in which the noise is explicitly estimated in the reproducing kernel Hilbert space. This enables KMNF dealing with non-linear relations between the noise and the signal features jointly. Results show that the proposed KMNF provides the most noise-free features when confronted...
Evaluating Clustering in Subspace Projections of High Dimensional Data
Müller, Emmanuel; Günnemann, Stephan; Assent, Ira
2009-01-01
Clustering high dimensional data is an emerging research field. Subspace clustering or projected clustering group similar objects in subspaces, i.e. projections, of the full space. In the past decade, several clustering paradigms have been developed in parallel, without thorough evaluation...... and comparison between these paradigms on a common basis. Conclusive evaluation and comparison is challenged by three major issues. First, there is no ground truth that describes the "true" clusters in real world data. Second, a large variety of evaluation measures have been used that reflect different aspects...... of the clustering result. Finally, in typical publications authors have limited their analysis to their favored paradigm only, while paying other paradigms little or no attention. In this paper, we take a systematic approach to evaluate the major paradigms in a common framework. We study representative clustering...
Invariant subspaces in some function spaces on symmetric spaces. II
Platonov, S S
1998-01-01
Let G be a semisimple connected Lie group with finite centre, K a maximal compact subgroup of G, and M=G/K a Riemannian symmetric space of non-compact type. We study the problem of describing the structure of closed linear subspaces in various function spaces on M that are invariant under the quasiregular representation of the group G. We consider the case when M is a symplectic symmetric space of rank 1
Quantum cloning of mixed states in symmetric subspaces
Fan Heng
2003-01-01
Quantum-cloning machine for arbitrary mixed states in symmetric subspaces is proposed. This quantum-cloning machine can be used to copy part of the output state of another quantum-cloning machine and is useful in quantum computation and quantum information. The shrinking factor of this quantum cloning achieves the well-known upper bound. When the input is identical pure states, two different fidelities of this cloning machine are optimal
Bi Sparsity Pursuit: A Paradigm for Robust Subspace Recovery
2016-09-27
Bian, Student Member, IEEE, and Hamid Krim, Fellow, IEEE Abstract The success of sparse models in computer vision and machine learning is due to the...16. SECURITY CLASSIFICATION OF: The success of sparse models in computer vision and machine learning is due to the fact that, high dimensional data...vision and machine learning is due to the fact that, high dimensional data is distributed in a union of low dimensional subspaces in many real-world
Index Formulae for Subspaces of Kreĭn Spaces
Dijksma, Aad; Gheondea, Aurelian
1996-01-01
For a subspace S of a Kreĭn space K and an arbitrary fundamental decomposition K = K-[+]K+ of K, we prove the index formula κ-(S) + dim(S⊥ ∩ K+) = κ+(S⊥) + dim(S ∩ K-), where κ±(S) stands for the positive/negative signature of S. The difference dim(S ∩ K-) - dim(S⊥ ∩ K+), provided it is well
An adaptation of Krylov subspace methods to path following
Walker, H.F. [Utah State Univ., Logan, UT (United States)
1996-12-31
Krylov subspace methods at present constitute a very well known and highly developed class of iterative linear algebra methods. These have been effectively applied to nonlinear system solving through Newton-Krylov methods, in which Krylov subspace methods are used to solve the linear systems that characterize steps of Newton`s method (the Newton equations). Here, we will discuss the application of Krylov subspace methods to path following problems, in which the object is to track a solution curve as a parameter varies. Path following methods are typically of predictor-corrector form, in which a point near the solution curve is {open_quotes}predicted{close_quotes} by some easy but relatively inaccurate means, and then a series of Newton-like corrector iterations is used to return approximately to the curve. The analogue of the Newton equation is underdetermined, and an additional linear condition must be specified to determine corrector steps uniquely. This is typically done by requiring that the steps be orthogonal to an approximate tangent direction. Augmenting the under-determined system with this orthogonality condition in a straightforward way typically works well if direct linear algebra methods are used, but Krylov subspace methods are often ineffective with this approach. We will discuss recent work in which this orthogonality condition is imposed directly as a constraint on the corrector steps in a certain way. The means of doing this preserves problem conditioning, allows the use of preconditioners constructed for the fixed-parameter case, and has certain other advantages. Experiments on standard PDE continuation test problems indicate that this approach is effective.
Bruemmer, David J [Idaho Falls, ID
2009-11-17
A robot platform includes perceptors, locomotors, and a system controller. The system controller executes a robot intelligence kernel (RIK) that includes a multi-level architecture and a dynamic autonomy structure. The multi-level architecture includes a robot behavior level for defining robot behaviors, that incorporate robot attributes and a cognitive level for defining conduct modules that blend an adaptive interaction between predefined decision functions and the robot behaviors. The dynamic autonomy structure is configured for modifying a transaction capacity between an operator intervention and a robot initiative and may include multiple levels with at least a teleoperation mode configured to maximize the operator intervention and minimize the robot initiative and an autonomous mode configured to minimize the operator intervention and maximize the robot initiative. Within the RIK at least the cognitive level includes the dynamic autonomy structure.
Comparison Study of Subspace Identification Methods Applied to Flexible Structures
Abdelghani, M.; Verhaegen, M.; Van Overschee, P.; De Moor, B.
1998-09-01
In the past few years, various time domain methods for identifying dynamic models of mechanical structures from modal experimental data have appeared. Much attention has been given recently to so-called subspace methods for identifying state space models. This paper presents a detailed comparison study of these subspace identification methods: the eigensystem realisation algorithm with observer/Kalman filter Markov parameters computed from input/output data (ERA/OM), the robust version of the numerical algorithm for subspace system identification (N4SID), and a refined version of the past outputs scheme of the multiple-output error state space (MOESP) family of algorithms. The comparison is performed by simulating experimental data using the five mode reduced model of the NASA Mini-Mast structure. The general conclusion is that for the case of white noise excitations as well as coloured noise excitations, the N4SID/MOESP algorithms perform equally well but give better results (improved transfer function estimates, improved estimates of the output) compared to the ERA/OM algorithm. The key computational step in the three algorithms is the approximation of the extended observability matrix of the system to be identified, for N4SID/MOESP, or of the observer for the system to be identified, for the ERA/OM. Furthermore, the three algorithms only require the specification of one dimensioning parameter.
Improved Stochastic Subspace System Identification for Structural Health Monitoring
Chang, Chia-Ming; Loh, Chin-Hsiung
2015-07-01
Structural health monitoring acquires structural information through numerous sensor measurements. Vibrational measurement data render the dynamic characteristics of structures to be extracted, in particular of the modal properties such as natural frequencies, damping, and mode shapes. The stochastic subspace system identification has been recognized as a power tool which can present a structure in the modal coordinates. To obtain qualitative identified data, this tool needs to spend computational expense on a large set of measurements. In study, a stochastic system identification framework is proposed to improve the efficiency and quality of the conventional stochastic subspace system identification. This framework includes 1) measured signal processing, 2) efficient space projection, 3) system order selection, and 4) modal property derivation. The measured signal processing employs the singular spectrum analysis algorithm to lower the noise components as well as to present a data set in a reduced dimension. The subspace is subsequently derived from the data set presented in a delayed coordinate. With the proposed order selection criteria, the number of structural modes is determined, resulting in the modal properties. This system identification framework is applied to a real-world bridge for exploring the feasibility in real-time applications. The results show that this improved system identification method significantly decreases computational time, while qualitative modal parameters are still attained.
Random Subspace Aggregation for Cancer Prediction with Gene Expression Profiles
Liying Yang
2016-01-01
Full Text Available Background. Precisely predicting cancer is crucial for cancer treatment. Gene expression profiles make it possible to analyze patterns between genes and cancers on the genome-wide scale. Gene expression data analysis, however, is confronted with enormous challenges for its characteristics, such as high dimensionality, small sample size, and low Signal-to-Noise Ratio. Results. This paper proposes a method, termed RS_SVM, to predict gene expression profiles via aggregating SVM trained on random subspaces. After choosing gene features through statistical analysis, RS_SVM randomly selects feature subsets to yield random subspaces and training SVM classifiers accordingly and then aggregates SVM classifiers to capture the advantage of ensemble learning. Experiments on eight real gene expression datasets are performed to validate the RS_SVM method. Experimental results show that RS_SVM achieved better classification accuracy and generalization performance in contrast with single SVM, K-nearest neighbor, decision tree, Bagging, AdaBoost, and the state-of-the-art methods. Experiments also explored the effect of subspace size on prediction performance. Conclusions. The proposed RS_SVM method yielded superior performance in analyzing gene expression profiles, which demonstrates that RS_SVM provides a good channel for such biological data.
The Flux OSKit: A Substrate for Kernel and Language Research
1997-10-01
unclassified Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 tions. Our own microkernel -based OS, Fluke [17], puts almost all of the OSKit to use...kernels distance the language from the hardware; even microkernels and other extensible kernels enforce some default policy which often conflicts with a...be particu- larly useful in these research projects. 6.1.1 The Fluke OS In 1996 we developed an entirely new microkernel - based system called Fluke
Visualization of nonlinear kernel models in neuroimaging by sensitivity maps
Rasmussen, Peter Mondrup; Hansen, Lars Kai; Madsen, Kristoffer Hougaard
There is significant current interest in decoding mental states from neuroimages. In this context kernel methods, e.g., support vector machines (SVM) are frequently adopted to learn statistical relations between patterns of brain activation and experimental conditions. In this paper we focus...... on visualization of such nonlinear kernel models. Specifically, we investigate the sensitivity map as a technique for generation of global summary maps of kernel classification methods. We illustrate the performance of the sensitivity map on functional magnetic resonance (fMRI) data based on visual stimuli. We...
Zhang, Xiaole; Efthimiou, George; Wang, Yan; Huang, Meng
2018-04-01
Radiation from the deposited radionuclides is indispensable information for environmental impact assessment of nuclear power plants and emergency management during nuclear accidents. Ground shine estimation is related to multiple physical processes, including atmospheric dispersion, deposition, soil and air radiation shielding. It still remains unclear that whether the normally adopted "infinite plane" source assumption for the ground shine calculation is accurate enough, especially for the area with highly heterogeneous deposition distribution near the release point. In this study, a new ground shine calculation scheme, which accounts for both the spatial deposition distribution and the properties of air and soil layers, is developed based on point kernel method. Two sets of "detector-centered" grids are proposed and optimized for both the deposition and radiation calculations to better simulate the results measured by the detectors, which will be beneficial for the applications such as source term estimation. The evaluation against the available data of Monte Carlo methods in the literature indicates that the errors of the new scheme are within 5% for the key radionuclides in nuclear accidents. The comparisons between the new scheme and "infinite plane" assumption indicate that the assumption is tenable (relative errors within 20%) for the area located 1 km away from the release source. Within 1 km range, the assumption mainly causes errors for wet deposition and the errors are independent of rain intensities. The results suggest that the new scheme should be adopted if the detectors are within 1 km from the source under the stable atmosphere (classes E and F), or the detectors are within 500 m under slightly unstable (class C) or neutral (class D) atmosphere. Otherwise, the infinite plane assumption is reasonable since the relative errors induced by this assumption are within 20%. The results here are only based on theoretical investigations. They should
Improved magnetic resonance fingerprinting reconstruction with low-rank and subspace modeling.
Zhao, Bo; Setsompop, Kawin; Adalsteinsson, Elfar; Gagoski, Borjan; Ye, Huihui; Ma, Dan; Jiang, Yun; Ellen Grant, P; Griswold, Mark A; Wald, Lawrence L
2018-02-01
This article introduces a constrained imaging method based on low-rank and subspace modeling to improve the accuracy and speed of MR fingerprinting (MRF). A new model-based imaging method is developed for MRF to reconstruct high-quality time-series images and accurate tissue parameter maps (e.g., T 1 , T 2 , and spin density maps). Specifically, the proposed method exploits low-rank approximations of MRF time-series images, and further enforces temporal subspace constraints to capture magnetization dynamics. This allows the time-series image reconstruction problem to be formulated as a simple linear least-squares problem, which enables efficient computation. After image reconstruction, tissue parameter maps are estimated via dictionary-based pattern matching, as in the conventional approach. The effectiveness of the proposed method was evaluated with in vivo experiments. Compared with the conventional MRF reconstruction, the proposed method reconstructs time-series images with significantly reduced aliasing artifacts and noise contamination. Although the conventional approach exhibits some robustness to these corruptions, the improved time-series image reconstruction in turn provides more accurate tissue parameter maps. The improvement is pronounced especially when the acquisition time becomes short. The proposed method significantly improves the accuracy of MRF, and also reduces data acquisition time. Magn Reson Med 79:933-942, 2018. © 2017 International Society for Magnetic Resonance in Medicine. © 2017 International Society for Magnetic Resonance in Medicine.
Per-Sample Multiple Kernel Approach for Visual Concept Learning
Ling-Yu Duan
2010-01-01
Full Text Available Learning visual concepts from images is an important yet challenging problem in computer vision and multimedia research areas. Multiple kernel learning (MKL methods have shown great advantages in visual concept learning. As a visual concept often exhibits great appearance variance, a canonical MKL approach may not generate satisfactory results when a uniform kernel combination is applied over the input space. In this paper, we propose a per-sample multiple kernel learning (PS-MKL approach to take into account intraclass diversity for improving discrimination. PS-MKL determines sample-wise kernel weights according to kernel functions and training samples. Kernel weights as well as kernel-based classifiers are jointly learned. For efficient learning, PS-MKL employs a sample selection strategy. Extensive experiments are carried out over three benchmarking datasets of different characteristics including Caltech101, WikipediaMM, and Pascal VOC'07. PS-MKL has achieved encouraging performance, comparable to the state of the art, which has outperformed a canonical MKL.
Per-Sample Multiple Kernel Approach for Visual Concept Learning
Tian Yonghong
2010-01-01
Full Text Available Abstract Learning visual concepts from images is an important yet challenging problem in computer vision and multimedia research areas. Multiple kernel learning (MKL methods have shown great advantages in visual concept learning. As a visual concept often exhibits great appearance variance, a canonical MKL approach may not generate satisfactory results when a uniform kernel combination is applied over the input space. In this paper, we propose a per-sample multiple kernel learning (PS-MKL approach to take into account intraclass diversity for improving discrimination. PS-MKL determines sample-wise kernel weights according to kernel functions and training samples. Kernel weights as well as kernel-based classifiers are jointly learned. For efficient learning, PS-MKL employs a sample selection strategy. Extensive experiments are carried out over three benchmarking datasets of different characteristics including Caltech101, WikipediaMM, and Pascal VOC'07. PS-MKL has achieved encouraging performance, comparable to the state of the art, which has outperformed a canonical MKL.
Mixture Density Mercer Kernels: A Method to Learn Kernels
National Aeronautics and Space Administration — This paper presents a method of generating Mercer Kernels from an ensemble of probabilistic mixture models, where each mixture model is generated from a Bayesian...
Drozdowicz, K.
1995-01-01
A comprehensive unified description of the application of Granada's Synthetic Model to the slow-neutron scattering by the molecular systems is continued. Detailed formulae for the zero-order energy transfer kernel are presented basing on the general formalism of the model. An explicit analytical formula for the total scattering cross section as a function of the incident neutron energy is also obtained. Expressions of the free gas model for the zero-order scattering kernel and for total scattering kernel are considered as a sub-case of the Synthetic Model. (author). 10 refs
Simultaneous multislice magnetic resonance fingerprinting with low-rank and subspace modeling.
Bo Zhao; Bilgic, Berkin; Adalsteinsson, Elfar; Griswold, Mark A; Wald, Lawrence L; Setsompop, Kawin
2017-07-01
Magnetic resonance fingerprinting (MRF) is a new quantitative imaging paradigm that enables simultaneous acquisition of multiple magnetic resonance tissue parameters (e.g., T 1 , T 2 , and spin density). Recently, MRF has been integrated with simultaneous multislice (SMS) acquisitions to enable volumetric imaging with faster scan time. In this paper, we present a new image reconstruction method based on low-rank and subspace modeling for improved SMS-MRF. Here the low-rank model exploits strong spatiotemporal correlation among contrast-weighted images, while the subspace model captures the temporal evolution of magnetization dynamics. With the proposed model, the image reconstruction problem is formulated as a convex optimization problem, for which we develop an algorithm based on variable splitting and the alternating direction method of multipliers. The performance of the proposed method has been evaluated by numerical experiments, and the results demonstrate that the proposed method leads to improved accuracy over the conventional approach. Practically, the proposed method has a potential to allow for a 3× speedup with minimal reconstruction error, resulting in less than 5 sec imaging time per slice.
Fan, J; Fan, J; Hu, W; Wang, J [Fudan University Shanghai Cancer Center, Shanghai, Shanghai (China)
2016-06-15
Purpose: To develop a fast automatic algorithm based on the two dimensional kernel density estimation (2D KDE) to predict the dose-volume histogram (DVH) which can be employed for the investigation of radiotherapy quality assurance and automatic treatment planning. Methods: We propose a machine learning method that uses previous treatment plans to predict the DVH. The key to the approach is the framing of DVH in a probabilistic setting. The training consists of estimating, from the patients in the training set, the joint probability distribution of the dose and the predictive features. The joint distribution provides an estimation of the conditional probability of the dose given the values of the predictive features. For the new patient, the prediction consists of estimating the distribution of the predictive features and marginalizing the conditional probability from the training over this. Integrating the resulting probability distribution for the dose yields an estimation of the DVH. The 2D KDE is implemented to predict the joint probability distribution of the training set and the distribution of the predictive features for the new patient. Two variables, including the signed minimal distance from each OAR (organs at risk) voxel to the target boundary and its opening angle with respect to the origin of voxel coordinate, are considered as the predictive features to represent the OAR-target spatial relationship. The feasibility of our method has been demonstrated with the rectum, breast and head-and-neck cancer cases by comparing the predicted DVHs with the planned ones. Results: The consistent result has been found between these two DVHs for each cancer and the average of relative point-wise differences is about 5% within the clinical acceptable extent. Conclusion: According to the result of this study, our method can be used to predict the clinical acceptable DVH and has ability to evaluate the quality and consistency of the treatment planning.
Fan, J; Fan, J; Hu, W; Wang, J
2016-01-01
Purpose: To develop a fast automatic algorithm based on the two dimensional kernel density estimation (2D KDE) to predict the dose-volume histogram (DVH) which can be employed for the investigation of radiotherapy quality assurance and automatic treatment planning. Methods: We propose a machine learning method that uses previous treatment plans to predict the DVH. The key to the approach is the framing of DVH in a probabilistic setting. The training consists of estimating, from the patients in the training set, the joint probability distribution of the dose and the predictive features. The joint distribution provides an estimation of the conditional probability of the dose given the values of the predictive features. For the new patient, the prediction consists of estimating the distribution of the predictive features and marginalizing the conditional probability from the training over this. Integrating the resulting probability distribution for the dose yields an estimation of the DVH. The 2D KDE is implemented to predict the joint probability distribution of the training set and the distribution of the predictive features for the new patient. Two variables, including the signed minimal distance from each OAR (organs at risk) voxel to the target boundary and its opening angle with respect to the origin of voxel coordinate, are considered as the predictive features to represent the OAR-target spatial relationship. The feasibility of our method has been demonstrated with the rectum, breast and head-and-neck cancer cases by comparing the predicted DVHs with the planned ones. Results: The consistent result has been found between these two DVHs for each cancer and the average of relative point-wise differences is about 5% within the clinical acceptable extent. Conclusion: According to the result of this study, our method can be used to predict the clinical acceptable DVH and has ability to evaluate the quality and consistency of the treatment planning.
Urrutia, Eugene; Lee, Seunggeun; Maity, Arnab; Zhao, Ni; Shen, Judong; Li, Yun; Wu, Michael C
Analysis of rare genetic variants has focused on region-based analysis wherein a subset of the variants within a genomic region is tested for association with a complex trait. Two important practical challenges have emerged. First, it is difficult to choose which test to use. Second, it is unclear which group of variants within a region should be tested. Both depend on the unknown true state of nature. Therefore, we develop the Multi-Kernel SKAT (MK-SKAT) which tests across a range of rare variant tests and groupings. Specifically, we demonstrate that several popular rare variant tests are special cases of the sequence kernel association test which compares pair-wise similarity in trait value to similarity in the rare variant genotypes between subjects as measured through a kernel function. Choosing a particular test is equivalent to choosing a kernel. Similarly, choosing which group of variants to test also reduces to choosing a kernel. Thus, MK-SKAT uses perturbation to test across a range of kernels. Simulations and real data analyses show that our framework controls type I error while maintaining high power across settings: MK-SKAT loses power when compared to the kernel for a particular scenario but has much greater power than poor choices.
2010-01-01
... 7 Agriculture 8 2010-01-01 2010-01-01 false Kernel weight. 981.9 Section 981.9 Agriculture Regulations of the Department of Agriculture (Continued) AGRICULTURAL MARKETING SERVICE (Marketing Agreements... Regulating Handling Definitions § 981.9 Kernel weight. Kernel weight means the weight of kernels, including...
2010-01-01
... 7 Agriculture 2 2010-01-01 2010-01-01 false Half kernel. 51.2295 Section 51.2295 Agriculture... Standards for Shelled English Walnuts (Juglans Regia) Definitions § 51.2295 Half kernel. Half kernel means the separated half of a kernel with not more than one-eighth broken off. ...
Mallika V
2010-03-01
Full Text Available Type III Polyketide synthases (PKS are family of proteins considered to have significant role in the biosynthesis of various polyketides in plants, fungi and bacteria. As these proteins show positive effects to human health, more researches are going on regarding this particular protein. Developing a tool to identify the probability of sequence, being a type III polyketide synthase will minimize the time consumption and manpower efforts. In this approach, we have designed and implemented PKSIIIpred, a high performance prediction server for type III PKS where the classifier is Support Vector Machine (SVM. Based on the limited training dataset, the tool efficiently predicts the type III PKS superfamily of proteins with high sensitivity and specificity. PKSIIIpred is available at http://type3pks.in/prediction/. We expect that this tool may serve as a useful resource for type III PKS researchers. Currently work is being progressed for further betterment of prediction accuracy by including more sequence features in the training dataset.
A kernel version of spatial factor analysis
Nielsen, Allan Aasbjerg
2009-01-01
. Schölkopf et al. introduce kernel PCA. Shawe-Taylor and Cristianini is an excellent reference for kernel methods in general. Bishop and Press et al. describe kernel methods among many other subjects. Nielsen and Canty use kernel PCA to detect change in univariate airborne digital camera images. The kernel...... version of PCA handles nonlinearities by implicitly transforming data into high (even infinite) dimensional feature space via the kernel function and then performing a linear analysis in that space. In this paper we shall apply kernel versions of PCA, maximum autocorrelation factor (MAF) analysis...
kernel oil by lipolytic organisms
USER
2010-08-02
Aug 2, 2010 ... Rancidity of extracted cashew oil was observed with cashew kernel stored at 70, 80 and 90% .... method of American Oil Chemist Society AOCS (1978) using glacial ..... changes occur and volatile products are formed that are.
A Kernel for Protein Secondary Structure Prediction
Guermeur , Yann; Lifchitz , Alain; Vert , Régis
2004-01-01
http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=10338&mode=toc; International audience; Multi-class support vector machines have already proved efficient in protein secondary structure prediction as ensemble methods, to combine the outputs of sets of classifiers based on different principles. In this chapter, their implementation as basic prediction methods, processing the primary structure or the profile of multiple alignments, is investigated. A kernel devoted to the task is in...
2017-09-27
100 times larger for the minimal Krylov subspace. 0 5 10 15 20 25 Krylov subspace dimension 10-2 10-1 100 101 102 103 104 jjĜ ¡ 1 jj F SVD...approximation Kn (G;u(0) ) 0 5 10 15 20 25 Krylov subspace dimension 10-2 10-1 100 101 102 103 104 jjx jj fo r m in x jjĜ x ¡ bjj SVD approximation Kn (G;u(0
Gamow state vectors as functionals over subspaces of the nuclear space
Bohm, A.
1979-12-01
Exponentially decaying Gamow state vectors are obtained from S-matrix poles in the lower half of the second sheet, and are defined as functionals over a subspace of the nuclear space, PHI. Exponentially growing Gamow state vectors are obtained from S-matrix poles in the upper half of the second sheet, and are defined as functionals over another subspace of PHI. On functionals over these two subspaces the dynamical group of time development splits into two semigroups
Multivariate and semiparametric kernel regression
Härdle, Wolfgang; Müller, Marlene
1997-01-01
The paper gives an introduction to theory and application of multivariate and semiparametric kernel smoothing. Multivariate nonparametric density estimation is an often used pilot tool for examining the structure of data. Regression smoothing helps in investigating the association between covariates and responses. We concentrate on kernel smoothing using local polynomial fitting which includes the Nadaraya-Watson estimator. Some theory on the asymptotic behavior and bandwidth selection is pro...
Barndorff-Nielsen, Ole E.
The density function of the gamma distribution is used as shift kernel in Brownian semistationary processes modelling the timewise behaviour of the velocity in turbulent regimes. This report presents exact and asymptotic properties of the second order structure function under such a model......, and relates these to results of von Karmann and Horwath. But first it is shown that the gamma kernel is interpretable as a Green’s function....
An efficient preconditioning technique using Krylov subspace methods for 3D characteristics solvers
Dahmani, M.; Le Tellier, R.; Roy, R.; Hebert, A.
2005-01-01
The Generalized Minimal RESidual (GMRES) method, using a Krylov subspace projection, is adapted and implemented to accelerate a 3D iterative transport solver based on the characteristics method. Another acceleration technique called the self-collision rebalancing technique (SCR) can also be used to accelerate the solution or as a left preconditioner for GMRES. The GMRES method is usually used to solve a linear algebraic system (Ax=b). It uses K(r (o) ,A) as projection subspace and AK(r (o) ,A) for the orthogonalization of the residual. This paper compares the performance of these two combined methods on various problems. To implement the GMRES iterative method, the characteristics equations are derived in linear algebra formalism by using the equivalence between the method of characteristics and the method of collision probability to end up with a linear algebraic system involving fluxes and currents. Numerical results show good performance of the GMRES technique especially for the cases presenting large material heterogeneity with a scattering ratio close to 1. Similarly, the SCR preconditioning slightly increases the GMRES efficiency
On the selection of user-defined parameters in data-driven stochastic subspace identification
Priori, C.; De Angelis, M.; Betti, R.
2018-02-01
The paper focuses on the time domain output-only technique called Data-Driven Stochastic Subspace Identification (DD-SSI); in order to identify modal models (frequencies, damping ratios and mode shapes), the role of its user-defined parameters is studied, and rules to determine their minimum values are proposed. Such investigation is carried out using, first, the time histories of structural responses to stationary excitations, with a large number of samples, satisfying the hypothesis on the input imposed by DD-SSI. Then, the case of non-stationary seismic excitations with a reduced number of samples is considered. In this paper, partitions of the data matrix different from the one proposed in the SSI literature are investigated, together with the influence of different choices of the weighting matrices. The study is carried out considering two different applications: (1) data obtained from vibration tests on a scaled structure and (2) in-situ tests on a reinforced concrete building. Referring to the former, the identification of a steel frame structure tested on a shaking table is performed using its responses in terms of absolute accelerations to a stationary (white noise) base excitation and to non-stationary seismic excitations of low intensity. Black-box and modal models are identified in both cases and the results are compared with those from an input-output subspace technique. With regards to the latter, the identification of a complex hospital building is conducted using data obtained from ambient vibration tests.
Renaut, R.; He, Q. [Arizona State Univ., Tempe, AZ (United States)
1994-12-31
In a new parallel iterative algorithm for unconstrained optimization by multisplitting is proposed. In this algorithm the original problem is split into a set of small optimization subproblems which are solved using well known sequential algorithms. These algorithms are iterative in nature, e.g. DFP variable metric method. Here the authors use sequential algorithms based on an inexact subspace search, which is an extension to the usual idea of an inexact fine search. Essentially the idea of the inexact line search for nonlinear minimization is that at each iteration the authors only find an approximate minimum in the line search direction. Hence by inexact subspace search, they mean that, instead of finding the minimum of the subproblem at each interation, they do an incomplete down hill search to give an approximate minimum. Some convergence and numerical results for this algorithm will be presented. Further, the original theory will be generalized to the situation with a singular Hessian. Applications for nonlinear least squares problems will be presented. Experimental results will be presented for implementations on an Intel iPSC/860 Hypercube with 64 nodes as well as on the Intel Paragon.
Subspace Analysis of Indoor UWB Channels
Rachid Saadane
2005-03-01
Full Text Available This work aims at characterizing the second-order statistics of indoor ultra-wideband (UWB channels using channel sounding techniques. We present measurement results for different scenarios conducted in a laboratory setting at Institut EurÃƒÂ©com. These are based on an eigendecomposition of the channel autocovariance matrix, which allows for the analysis of the growth in the number of significant degrees of freedom of the channel process as a function of the signaling bandwidth as well as the statistical correlation between different propagation paths. We show empirical eigenvalue distributions as a function of the signal bandwidth for both line-of-sight and non-line-of-sight situations. Furthermore, we give examples where paths from different propagation clusters (possibly arising from reflection or diffraction show strong statistical dependence.
Analytic scattering kernels for neutron thermalization studies
Sears, V.F.
1990-01-01
Current plans call for the inclusion of a liquid hydrogen or deuterium cold source in the NRU replacement vessel. This report is part of an ongoing study of neutron thermalization in such a cold source. Here, we develop a simple analytical model for the scattering kernel of monatomic and diatomic liquids. We also present the results of extensive numerical calculations based on this model for liquid hydrogen, liquid deuterium, and mixtures of the two. These calculations demonstrate the dependence of the scattering kernel on the incident and scattered-neutron energies, the behavior near rotational thresholds, the dependence on the centre-of-mass pair correlations, the dependence on the ortho concentration, and the dependence on the deuterium concentration in H 2 /D 2 mixtures. The total scattering cross sections are also calculated and compared with available experimental results
Quantized kernel least mean square algorithm.
Chen, Badong; Zhao, Songlin; Zhu, Pingping; Príncipe, José C
2012-01-01
In this paper, we propose a quantization approach, as an alternative of sparsification, to curb the growth of the radial basis function structure in kernel adaptive filtering. The basic idea behind this method is to quantize and hence compress the input (or feature) space. Different from sparsification, the new approach uses the "redundant" data to update the coefficient of the closest center. In particular, a quantized kernel least mean square (QKLMS) algorithm is developed, which is based on a simple online vector quantization method. The analytical study of the mean square convergence has been carried out. The energy conservation relation for QKLMS is established, and on this basis we arrive at a sufficient condition for mean square convergence, and a lower and upper bound on the theoretical value of the steady-state excess mean square error. Static function estimation and short-term chaotic time-series prediction examples are presented to demonstrate the excellent performance.
A kernel adaptive algorithm for quaternion-valued inputs.
Paul, Thomas K; Ogunfunmi, Tokunbo
2015-10-01
The use of quaternion data can provide benefit in applications like robotics and image recognition, and particularly for performing transforms in 3-D space. Here, we describe a kernel adaptive algorithm for quaternions. A least mean square (LMS)-based method was used, resulting in the derivation of the quaternion kernel LMS (Quat-KLMS) algorithm. Deriving this algorithm required describing the idea of a quaternion reproducing kernel Hilbert space (RKHS), as well as kernel functions suitable with quaternions. A modified HR calculus for Hilbert spaces was used to find the gradient of cost functions defined on a quaternion RKHS. In addition, the use of widely linear (or augmented) filtering is proposed to improve performance. The benefit of the Quat-KLMS and widely linear forms in learning nonlinear transformations of quaternion data are illustrated with simulations.
Linear and kernel methods for multivariate change detection
Canty, Morton J.; Nielsen, Allan Aasbjerg
2012-01-01
), as well as maximum autocorrelation factor (MAF) and minimum noise fraction (MNF) analyses of IR-MAD images, both linear and kernel-based (nonlinear), may further enhance change signals relative to no-change background. IDL (Interactive Data Language) implementations of IR-MAD, automatic radiometric...... normalization, and kernel PCA/MAF/MNF transformations are presented that function as transparent and fully integrated extensions of the ENVI remote sensing image analysis environment. The train/test approach to kernel PCA is evaluated against a Hebbian learning procedure. Matlab code is also available...... that allows fast data exploration and experimentation with smaller datasets. New, multiresolution versions of IR-MAD that accelerate convergence and that further reduce no-change background noise are introduced. Computationally expensive matrix diagonalization and kernel image projections are programmed...
A Krylov Subspace Method for Unstructured Mesh SN Transport Computation
Yoo, Han Jong; Cho, Nam Zin; Kim, Jong Woon; Hong, Ser Gi; Lee, Young Ouk
2010-01-01
Hong, et al., have developed a computer code MUST (Multi-group Unstructured geometry S N Transport) for the neutral particle transport calculations in three-dimensional unstructured geometry. In this code, the discrete ordinates transport equation is solved by using the discontinuous finite element method (DFEM) or the subcell balance methods with linear discontinuous expansion. In this paper, the conventional source iteration in the MUST code is replaced by the Krylov subspace method to reduce computing time and the numerical test results are given
Perturbation for Frames for a Subspace of a Hilbert Space
Christensen, Ole; deFlicht, C.; Lennard, C.
1997-01-01
We extend a classical result stating that a sufficiently small perturbation$\\{ g_i \\}$ of a Riesz sequence $\\{ f_i \\}$ in a Hilbert space $H$ is again a Riesz sequence. It turns out that the analog result for a frame does not holdunless the frame is complete. However, we are able to prove a very...... similarresult for frames in the case where the gap between the subspaces$\\overline{span} \\{f_i \\}$ and $\\overline{span} \\{ g_i \\}$ is small enough. We give a geometric interpretation of the result....
A new discrete dipole kernel for quantitative susceptibility mapping.
Milovic, Carlos; Acosta-Cabronero, Julio; Pinto, José Miguel; Mattern, Hendrik; Andia, Marcelo; Uribe, Sergio; Tejos, Cristian
2018-09-01
Most approaches for quantitative susceptibility mapping (QSM) are based on a forward model approximation that employs a continuous Fourier transform operator to solve a differential equation system. Such formulation, however, is prone to high-frequency aliasing. The aim of this study was to reduce such errors using an alternative dipole kernel formulation based on the discrete Fourier transform and discrete operators. The impact of such an approach on forward model calculation and susceptibility inversion was evaluated in contrast to the continuous formulation both with synthetic phantoms and in vivo MRI data. The discrete kernel demonstrated systematically better fits to analytic field solutions, and showed less over-oscillations and aliasing artifacts while preserving low- and medium-frequency responses relative to those obtained with the continuous kernel. In the context of QSM estimation, the use of the proposed discrete kernel resulted in error reduction and increased sharpness. This proof-of-concept study demonstrated that discretizing the dipole kernel is advantageous for QSM. The impact on small or narrow structures such as the venous vasculature might by particularly relevant to high-resolution QSM applications with ultra-high field MRI - a topic for future investigations. The proposed dipole kernel has a straightforward implementation to existing QSM routines. Copyright © 2018 Elsevier Inc. All rights reserved.
Wu, Xiao; Shen, Jiong; Li, Yiguo; Lee, Kwang Y
2014-05-01
This paper develops a novel data-driven fuzzy modeling strategy and predictive controller for boiler-turbine unit using fuzzy clustering and subspace identification (SID) methods. To deal with the nonlinear behavior of boiler-turbine unit, fuzzy clustering is used to provide an appropriate division of the operation region and develop the structure of the fuzzy model. Then by combining the input data with the corresponding fuzzy membership functions, the SID method is extended to extract the local state-space model parameters. Owing to the advantages of the both methods, the resulting fuzzy model can represent the boiler-turbine unit very closely, and a fuzzy model predictive controller is designed based on this model. As an alternative approach, a direct data-driven fuzzy predictive control is also developed following the same clustering and subspace methods, where intermediate subspace matrices developed during the identification procedure are utilized directly as the predictor. Simulation results show the advantages and effectiveness of the proposed approach. Copyright © 2014 ISA. Published by Elsevier Ltd. All rights reserved.
Influence Function and Robust Variant of Kernel Canonical Correlation Analysis
Alam, Md. Ashad; Fukumizu, Kenji; Wang, Yu-Ping
2017-01-01
Many unsupervised kernel methods rely on the estimation of the kernel covariance operator (kernel CO) or kernel cross-covariance operator (kernel CCO). Both kernel CO and kernel CCO are sensitive to contaminated data, even when bounded positive definite kernels are used. To the best of our knowledge, there are few well-founded robust kernel methods for statistical unsupervised learning. In addition, while the influence function (IF) of an estimator can characterize its robustness, asymptotic ...
Fabian Horst
Full Text Available Traditionally, gait analysis has been centered on the idea of average behavior and normality. On one hand, clinical diagnoses and therapeutic interventions typically assume that average gait patterns remain constant over time. On the other hand, it is well known that all our movements are accompanied by a certain amount of variability, which does not allow us to make two identical steps. The purpose of this study was to examine changes in the intra-individual gait patterns across different time-scales (i.e., tens-of-mins, tens-of-hours.Nine healthy subjects performed 15 gait trials at a self-selected speed on 6 sessions within one day (duration between two subsequent sessions from 10 to 90 mins. For each trial, time-continuous ground reaction forces and lower body joint angles were measured. A supervised learning model using a kernel-based discriminant regression was applied for classifying sessions within individual gait patterns.Discernable characteristics of intra-individual gait patterns could be distinguished between repeated sessions by classification rates of 67.8 ± 8.8% and 86.3 ± 7.9% for the six-session-classification of ground reaction forces and lower body joint angles, respectively. Furthermore, the one-on-one-classification showed that increasing classification rates go along with increasing time durations between two sessions and indicate that changes of gait patterns appear at different time-scales.Discernable characteristics between repeated sessions indicate continuous intrinsic changes in intra-individual gait patterns and suggest a predominant role of deterministic processes in human motor control and learning. Natural changes of gait patterns without any externally induced injury or intervention may reflect continuous adaptations of the motor system over several time-scales. Accordingly, the modelling of walking by means of average gait patterns that are assumed to be near constant over time needs to be reconsidered in the
Horst, Fabian; Eekhoff, Alexander; Newell, Karl M; Schöllhorn, Wolfgang I
2017-01-01
Traditionally, gait analysis has been centered on the idea of average behavior and normality. On one hand, clinical diagnoses and therapeutic interventions typically assume that average gait patterns remain constant over time. On the other hand, it is well known that all our movements are accompanied by a certain amount of variability, which does not allow us to make two identical steps. The purpose of this study was to examine changes in the intra-individual gait patterns across different time-scales (i.e., tens-of-mins, tens-of-hours). Nine healthy subjects performed 15 gait trials at a self-selected speed on 6 sessions within one day (duration between two subsequent sessions from 10 to 90 mins). For each trial, time-continuous ground reaction forces and lower body joint angles were measured. A supervised learning model using a kernel-based discriminant regression was applied for classifying sessions within individual gait patterns. Discernable characteristics of intra-individual gait patterns could be distinguished between repeated sessions by classification rates of 67.8 ± 8.8% and 86.3 ± 7.9% for the six-session-classification of ground reaction forces and lower body joint angles, respectively. Furthermore, the one-on-one-classification showed that increasing classification rates go along with increasing time durations between two sessions and indicate that changes of gait patterns appear at different time-scales. Discernable characteristics between repeated sessions indicate continuous intrinsic changes in intra-individual gait patterns and suggest a predominant role of deterministic processes in human motor control and learning. Natural changes of gait patterns without any externally induced injury or intervention may reflect continuous adaptations of the motor system over several time-scales. Accordingly, the modelling of walking by means of average gait patterns that are assumed to be near constant over time needs to be reconsidered in the context of
Colliandre, Lionel; Le Guilloux, Vincent; Bourg, Stephane; Morin-Allory, Luc
2012-02-27
High Throughput Screening (HTS) is a standard technique widely used to find hit compounds in drug discovery projects. The high costs associated with such experiments have highlighted the need to carefully design screening libraries in order to avoid wasting resources. Molecular diversity is an established concept that has been used to this end for many years. In this article, a new approach to quantify the molecular diversity of screening libraries is presented. The approach is based on the Delimited Reference Chemical Subspace (DRCS) methodology, a new method that can be used to delimit the densest subspace spanned by a reference library in a reduced 2D continuous space. A total of 22 diversity indices were implemented or adapted to this methodology, which is used here to remove outliers and obtain a relevant cell-based partition of the subspace. The behavior of these indices was assessed and compared in various extreme situations and with respect to a set of theoretical rules that a diversity function should satisfy when libraries of different sizes have to be compared. Some gold standard indices are found inappropriate in such a context, while none of the tested indices behave perfectly in all cases. Five DRCS-based indices accounting for different aspects of diversity were finally selected, and a simple framework is proposed to use them effectively. Various libraries have been profiled with respect to more specific subspaces, which further illustrate the interest of the method.
Integral equations with contrasting kernels
Theodore Burton
2008-01-01
Full Text Available In this paper we study integral equations of the form $x(t=a(t-\\int^t_0 C(t,sx(sds$ with sharply contrasting kernels typified by $C^*(t,s=\\ln (e+(t-s$ and $D^*(t,s=[1+(t-s]^{-1}$. The kernel assigns a weight to $x(s$ and these kernels have exactly opposite effects of weighting. Each type is well represented in the literature. Our first project is to show that for $a\\in L^2[0,\\infty$, then solutions are largely indistinguishable regardless of which kernel is used. This is a surprise and it leads us to study the essential differences. In fact, those differences become large as the magnitude of $a(t$ increases. The form of the kernel alone projects necessary conditions concerning the magnitude of $a(t$ which could result in bounded solutions. Thus, the next project is to determine how close we can come to proving that the necessary conditions are also sufficient. The third project is to show that solutions will be bounded for given conditions on $C$ regardless of whether $a$ is chosen large or small; this is important in real-world problems since we would like to have $a(t$ as the sum of a bounded, but badly behaved function, and a large well behaved function.
Beamforming using subspace estimation from a diagonally averaged sample covariance.
Quijano, Jorge E; Zurk, Lisa M
2017-08-01
The potential benefit of a large-aperture sonar array for high resolution target localization is often challenged by the lack of sufficient data required for adaptive beamforming. This paper introduces a Toeplitz-constrained estimator of the clairvoyant signal covariance matrix corresponding to multiple far-field targets embedded in background isotropic noise. The estimator is obtained by averaging along subdiagonals of the sample covariance matrix, followed by covariance extrapolation using the method of maximum entropy. The sample covariance is computed from limited data snapshots, a situation commonly encountered with large-aperture arrays in environments characterized by short periods of local stationarity. Eigenvectors computed from the Toeplitz-constrained covariance are used to construct signal-subspace projector matrices, which are shown to reduce background noise and improve detection of closely spaced targets when applied to subspace beamforming. Monte Carlo simulations corresponding to increasing array aperture suggest convergence of the proposed projector to the clairvoyant signal projector, thereby outperforming the classic projector obtained from the sample eigenvectors. Beamforming performance of the proposed method is analyzed using simulated data, as well as experimental data from the Shallow Water Array Performance experiment.
An Ensemble Approach to Building Mercer Kernels with Prior Information
Srivastava, Ashok N.; Schumann, Johann; Fischer, Bernd
2005-01-01
This paper presents a new methodology for automatic knowledge driven data mining based on the theory of Mercer Kernels, which are highly nonlinear symmetric positive definite mappings from the original image space to a very high, possibly dimensional feature space. we describe a new method called Mixture Density Mercer Kernels to learn kernel function directly from data, rather than using pre-defined kernels. These data adaptive kernels can encode prior knowledge in the kernel using a Bayesian formulation, thus allowing for physical information to be encoded in the model. Specifically, we demonstrate the use of the algorithm in situations with extremely small samples of data. We compare the results with existing algorithms on data from the Sloan Digital Sky Survey (SDSS) and demonstrate the method's superior performance against standard methods. The code for these experiments has been generated with the AUTOBAYES tool, which automatically generates efficient and documented C/C++ code from abstract statistical model specifications. The core of the system is a schema library which contains templates for learning and knowledge discovery algorithms like different versions of EM, or numeric optimization methods like conjugate gradient methods. The template instantiation is supported by symbolic-algebraic computations, which allows AUTOBAYES to find closed-form solutions and, where possible, to integrate them into the code.
Wawro, Megan; Sweeney, George F.; Rabin, Jeffrey M.
2011-01-01
This paper reports on a study investigating students' ways of conceptualizing key ideas in linear algebra, with the particular results presented here focusing on student interactions with the notion of subspace. In interviews conducted with eight undergraduates, we found students' initial descriptions of subspace often varied substantially from…
Multineuron spike train analysis with R-convolution linear combination kernel.
Tezuka, Taro
2018-06-01
A spike train kernel provides an effective way of decoding information represented by a spike train. Some spike train kernels have been extended to multineuron spike trains, which are simultaneously recorded spike trains obtained from multiple neurons. However, most of these multineuron extensions were carried out in a kernel-specific manner. In this paper, a general framework is proposed for extending any single-neuron spike train kernel to multineuron spike trains, based on the R-convolution kernel. Special subclasses of the proposed R-convolution linear combination kernel are explored. These subclasses have a smaller number of parameters and make optimization tractable when the size of data is limited. The proposed kernel was evaluated using Gaussian process regression for multineuron spike trains recorded from an animal brain. It was compared with the sum kernel and the population Spikernel, which are existing ways of decoding multineuron spike trains using kernels. The results showed that the proposed approach performs better than these kernels and also other commonly used neural decoding methods. Copyright © 2018 Elsevier Ltd. All rights reserved.
Walder, Christian; Henao, Ricardo; Mørup, Morten
We present three generalisations of Kernel Principal Components Analysis (KPCA) which incorporate knowledge of the class labels of a subset of the data points. The first, MV-KPCA, penalises within class variances similar to Fisher discriminant analysis. The second, LSKPCA is a hybrid of least...... squares regression and kernel PCA. The final LR-KPCA is an iteratively reweighted version of the previous which achieves a sigmoid loss function on the labeled points. We provide a theoretical risk bound as well as illustrative experiments on real and toy data sets....
Dieu Tien Bui
2016-04-01
Full Text Available The Cat Ba National Park area (Vietnam with its tropical forest is recognized as being part of the world biodiversity conservation by the United Nations Educational, Scientific and Cultural Organization (UNESCO and is a well-known destination for tourists, with around 500,000 travelers per year. This area has been the site for many research projects; however, no project has been carried out for forest fire susceptibility assessment. Thus, protection of the forest including fire prevention is one of the main concerns of the local authorities. This work aims to produce a tropical forest fire susceptibility map for the Cat Ba National Park area, which may be helpful for the local authorities in forest fire protection management. To obtain this purpose, first, historical forest fires and related factors were collected from various sources to construct a GIS database. Then, a forest fire susceptibility model was developed using Kernel logistic regression. The quality of the model was assessed using the Receiver Operating Characteristic (ROC curve, area under the ROC curve (AUC, and five statistical evaluation measures. The usability of the resulting model is further compared with a benchmark model, the support vector machine (SVM. The results show that the Kernel logistic regression model has a high level of performance in both the training and validation dataset, with a prediction capability of 92.2%. Since the Kernel logistic regression model outperforms the benchmark model, we conclude that the proposed model is a promising alternative tool that should also be considered for forest fire susceptibility mapping in other areas. The results of this study are useful for the local authorities in forest planning and management.
Protein fold recognition using geometric kernel data fusion.
Zakeri, Pooya; Jeuris, Ben; Vandebril, Raf; Moreau, Yves
2014-07-01
Various approaches based on features extracted from protein sequences and often machine learning methods have been used in the prediction of protein folds. Finding an efficient technique for integrating these different protein features has received increasing attention. In particular, kernel methods are an interesting class of techniques for integrating heterogeneous data. Various methods have been proposed to fuse multiple kernels. Most techniques for multiple kernel learning focus on learning a convex linear combination of base kernels. In addition to the limitation of linear combinations, working with such approaches could cause a loss of potentially useful information. We design several techniques to combine kernel matrices by taking more involved, geometry inspired means of these matrices instead of convex linear combinations. We consider various sequence-based protein features including information extracted directly from position-specific scoring matrices and local sequence alignment. We evaluate our methods for classification on the SCOP PDB-40D benchmark dataset for protein fold recognition. The best overall accuracy on the protein fold recognition test set obtained by our methods is ∼ 86.7%. This is an improvement over the results of the best existing approach. Moreover, our computational model has been developed by incorporating the functional domain composition of proteins through a hybridization model. It is observed that by using our proposed hybridization model, the protein fold recognition accuracy is further improved to 89.30%. Furthermore, we investigate the performance of our approach on the protein remote homology detection problem by fusing multiple string kernels. The MATLAB code used for our proposed geometric kernel fusion frameworks are publicly available at http://people.cs.kuleuven.be/∼raf.vandebril/homepage/software/geomean.php?menu=5/. © The Author 2014. Published by Oxford University Press.
Alexander Caicedo
2016-11-01
Full Text Available Clinical data is comprised by a large number of synchronously collected biomedical signals that are measured at different locations. Deciphering the interrelationships of these signals can yield important information about their dependence providing some useful clinical diagnostic data. For instance, by computing the coupling between Near-Infrared Spectroscopy signals (NIRS and systemic variables the status of the hemodynamic regulation mechanisms can be assessed. In this paper we introduce an algorithm for the decomposition of NIRS signals into additive components. The algorithm, SIgnal DEcomposition base on Obliques Subspace Projections (SIDE-ObSP, assumes that the measured NIRS signal is a linear combination of the systemic measurements, following the linear regression model y = Ax + _. SIDE-ObSP decomposes the output such that, each component in the decomposition represents the sole linear influence of one corresponding regressor variable. This decomposition scheme aims at providing a better understanding of the relation between NIRS and systemic variables, and to provide a framework for the clinical interpretation of regression algorithms, thereby, facilitating their introduction into clinical practice. SIDE-ObSP combines oblique subspace projections (ObSP with the structure of a mean average system in order to define adequate signal subspaces. To guarantee smoothness in the estimated regression parameters, as observed in normal physiological processes, we impose a Tikhonov regularization using a matrix differential operator. We evaluate the performance of SIDE-ObSP by using a synthetic dataset, and present two case studies in the field of cerebral hemodynamics monitoring using NIRS. In addition, we compare the performance of this method with other system identification techniques. In the first case study data from 20 neonates during the first three days of life was used, here SIDE-ObSP decoupled the influence of changes in arterial oxygen
Caicedo, Alexander; Varon, Carolina; Hunyadi, Borbala; Papademetriou, Maria; Tachtsidis, Ilias; Van Huffel, Sabine
2016-01-01
Clinical data is comprised by a large number of synchronously collected biomedical signals that are measured at different locations. Deciphering the interrelationships of these signals can yield important information about their dependence providing some useful clinical diagnostic data. For instance, by computing the coupling between Near-Infrared Spectroscopy signals (NIRS) and systemic variables the status of the hemodynamic regulation mechanisms can be assessed. In this paper we introduce an algorithm for the decomposition of NIRS signals into additive components. The algorithm, SIgnal DEcomposition base on Obliques Subspace Projections (SIDE-ObSP), assumes that the measured NIRS signal is a linear combination of the systemic measurements, following the linear regression model y = Ax + ϵ . SIDE-ObSP decomposes the output such that, each component in the decomposition represents the sole linear influence of one corresponding regressor variable. This decomposition scheme aims at providing a better understanding of the relation between NIRS and systemic variables, and to provide a framework for the clinical interpretation of regression algorithms, thereby, facilitating their introduction into clinical practice. SIDE-ObSP combines oblique subspace projections (ObSP) with the structure of a mean average system in order to define adequate signal subspaces. To guarantee smoothness in the estimated regression parameters, as observed in normal physiological processes, we impose a Tikhonov regularization using a matrix differential operator. We evaluate the performance of SIDE-ObSP by using a synthetic dataset, and present two case studies in the field of cerebral hemodynamics monitoring using NIRS. In addition, we compare the performance of this method with other system identification techniques. In the first case study data from 20 neonates during the first 3 days of life was used, here SIDE-ObSP decoupled the influence of changes in arterial oxygen saturation from the
Multiple Kernel Learning with Data Augmentation
2016-11-22
JMLR: Workshop and Conference Proceedings 63:49–64, 2016 ACML 2016 Multiple Kernel Learning with Data Augmentation Khanh Nguyen nkhanh@deakin.edu.au...University, Australia Editors: Robert J. Durrant and Kee-Eung Kim Abstract The motivations of multiple kernel learning (MKL) approach are to increase... kernel expres- siveness capacity and to avoid the expensive grid search over a wide spectrum of kernels . A large amount of work has been proposed to
Sun, L.G.; De Visser, C.C.; Chu, Q.P.; Mulder, J.A.
2012-01-01
The optimality of the kernel number and kernel centers plays a significant role in determining the approximation power of nearly all kernel methods. However, the process of choosing optimal kernels is always formulated as a global optimization task, which is hard to accomplish. Recently, an