International Nuclear Information System (INIS)
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.
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.
Parameter optimization in the regularized kernel minimum noise fraction transformation
DEFF Research Database (Denmark)
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....
DEFF Research Database (Denmark)
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...
Multiple Kernel Learning for adaptive graph regularized nonnegative matrix factorization
Wang, Jim Jing-Yan; AbdulJabbar, Mustafa Abdulmajeed
2012-01-01
Nonnegative Matrix Factorization (NMF) has been continuously evolving in several areas like pattern recognition and information retrieval methods. It factorizes a matrix into a product of 2 low-rank non-negative matrices that will define parts-based, and linear representation of non-negative data. Recently, Graph regularized NMF (GrNMF) is proposed to find a compact representation, which uncovers the hidden semantics and simultaneously respects the intrinsic geometric structure. In GNMF, an affinity graph is constructed from the original data space to encode the geometrical information. In this paper, we propose a novel idea which engages a Multiple Kernel Learning approach into refining the graph structure that reflects the factorization of the matrix and the new data space. The GrNMF is improved by utilizing the graph refined by the kernel learning, and then a novel kernel learning method is introduced under the GrNMF framework. Our approach shows encouraging results of the proposed algorithm in comparison to the state-of-the-art clustering algorithms like NMF, GrNMF, SVD etc.
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...
Wang, Jim Jing-Yan
2014-09-20
Nonnegative matrix factorization (NMF), a popular part-based representation technique, does not capture the intrinsic local geometric structure of the data space. Graph regularized NMF (GNMF) was recently proposed to avoid this limitation by regularizing NMF with a nearest neighbor graph constructed from the input data set. However, GNMF has two main bottlenecks. First, using the original feature space directly to construct the graph is not necessarily optimal because of the noisy and irrelevant features and nonlinear distributions of data samples. Second, one possible way to handle the nonlinear distribution of data samples is by kernel embedding. However, it is often difficult to choose the most suitable kernel. To solve these bottlenecks, we propose two novel graph-regularized NMF methods, AGNMFFS and AGNMFMK, by introducing feature selection and multiple-kernel learning to the graph regularized NMF, respectively. Instead of using a fixed graph as in GNMF, the two proposed methods learn the nearest neighbor graph that is adaptive to the selected features and learned multiple kernels, respectively. For each method, we propose a unified objective function to conduct feature selection/multi-kernel learning, NMF and adaptive graph regularization simultaneously. We further develop two iterative algorithms to solve the two optimization problems. Experimental results on two challenging pattern classification tasks demonstrate that the proposed methods significantly outperform state-of-the-art data representation methods.
Ma, Qian; Xia, Houping; Xu, Qiang; Zhao, Lei
2018-05-01
A new method combining Tikhonov regularization and kernel matrix optimization by multi-wavelength incidence is proposed for retrieving particle size distribution (PSD) in an independent model with improved accuracy and stability. In comparison to individual regularization or multi-wavelength least squares, the proposed method exhibited better anti-noise capability, higher accuracy and stability. While standard regularization typically makes use of the unit matrix, it is not universal for different PSDs, particularly for Junge distributions. Thus, a suitable regularization matrix was chosen by numerical simulation, with the second-order differential matrix found to be appropriate for most PSD types.
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...
Natalia Y Bilenko; Jack L Gallant; Jack L Gallant
2016-01-01
In this article we introduce Pyrcca, an open-source Python package for performing canonical correlation analysis (CCA). CCA is a multivariate analysis method for identifying relationships between sets of variables. Pyrcca supports CCA with or without regularization, and with or without linear, polynomial, or Gaussian kernelization. We first use an abstract example to describe Pyrcca functionality. We then demonstrate how Pyrcca can be used to analyze neuroimaging data. Specifically, we use Py...
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...
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...
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
Digital signal processing with kernel methods
Rojo-Alvarez, José Luis; Muñoz-Marí, Jordi; Camps-Valls, Gustavo
2018-01-01
A realistic and comprehensive review of joint approaches to machine learning and signal processing algorithms, with application to communications, multimedia, and biomedical engineering systems Digital Signal Processing with Kernel Methods reviews the milestones in the mixing of classical digital signal processing models and advanced kernel machines statistical learning tools. It explains the fundamental concepts from both fields of machine learning and signal processing so that readers can quickly get up to speed in order to begin developing the concepts and application software in their own research. Digital Signal Processing with Kernel Methods provides a comprehensive overview of kernel methods in signal processing, without restriction to any application field. It also offers example applications and detailed benchmarking experiments with real and synthetic datasets throughout. Readers can find further worked examples with Matlab source code on a website developed by the authors. * Presents the necess...
The gravitational potential of axially symmetric bodies from a regularized green kernel
Trova, A.; Huré, J.-M.; Hersant, F.
2011-12-01
The determination of the gravitational potential inside celestial bodies (rotating stars, discs, planets, asteroids) is a common challenge in numerical Astrophysics. Under axial symmetry, the potential is classically found from a two-dimensional integral over the body's meridional cross-section. Because it involves an improper integral, high accuracy is generally difficult to reach. We have discovered that, for homogeneous bodies, the singular Green kernel can be converted into a regular kernel by direct analytical integration. This new kernel, easily managed with standard techniques, opens interesting horizons, not only for numerical calculus but also to generate approximations, in particular for geometrically thin discs and rings.
Kan, Hirohito; Kasai, Harumasa; Arai, Nobuyuki; Kunitomo, Hiroshi; Hirose, Yasujiro; Shibamoto, Yuta
2016-09-01
An effective background field removal technique is desired for more accurate quantitative susceptibility mapping (QSM) prior to dipole inversion. The aim of this study was to evaluate the accuracy of regularization enabled sophisticated harmonic artifact reduction for phase data with varying spherical kernel sizes (REV-SHARP) method using a three-dimensional head phantom and human brain data. The proposed REV-SHARP method used the spherical mean value operation and Tikhonov regularization in the deconvolution process, with varying 2-14mm kernel sizes. The kernel sizes were gradually reduced, similar to the SHARP with varying spherical kernel (VSHARP) method. We determined the relative errors and relationships between the true local field and estimated local field in REV-SHARP, VSHARP, projection onto dipole fields (PDF), and regularization enabled SHARP (RESHARP). Human experiment was also conducted using REV-SHARP, VSHARP, PDF, and RESHARP. The relative errors in the numerical phantom study were 0.386, 0.448, 0.838, and 0.452 for REV-SHARP, VSHARP, PDF, and RESHARP. REV-SHARP result exhibited the highest correlation between the true local field and estimated local field. The linear regression slopes were 1.005, 1.124, 0.988, and 0.536 for REV-SHARP, VSHARP, PDF, and RESHARP in regions of interest on the three-dimensional head phantom. In human experiments, no obvious errors due to artifacts were present in REV-SHARP. The proposed REV-SHARP is a new method combined with variable spherical kernel size and Tikhonov regularization. This technique might make it possible to be more accurate backgroud field removal and help to achive better accuracy of QSM. Copyright © 2016 Elsevier Inc. All rights reserved.
Dense Medium Machine Processing Method for Palm Kernel/ Shell ...
African Journals Online (AJOL)
ADOWIE PERE
Cracked palm kernel is a mixture of kernels, broken shells, dusts and other impurities. In ... machine processing method using dense medium, a separator, a shell collector and a kernel .... efficiency, ease of maintenance and uniformity of.
Directory of Open Access Journals (Sweden)
Natalia Y Bilenko
2016-11-01
Full Text Available In this article we introduce Pyrcca, an open-source Python package for performing canonical correlation analysis (CCA. CCA is a multivariate analysis method for identifying relationships between sets of variables. Pyrcca supports CCA with or without regularization, and with or without linear, polynomial, or Gaussian kernelization. We first use an abstract example to describe Pyrcca functionality. We then demonstrate how Pyrcca can be used to analyze neuroimaging data. Specifically, we use Pyrcca to implement cross-subject comparison in a natural movie functional magnetic resonance imaging (fMRI experiment by finding a data-driven set of functional response patterns that are similar across individuals. We validate this cross-subject comparison method in Pyrcca by predicting responses to novel natural movies across subjects. Finally, we show how Pyrcca can reveal retinotopic organization in brain responses to natural movies without the need for an explicit model.
Bilenko, Natalia Y; Gallant, Jack L
2016-01-01
In this article we introduce Pyrcca, an open-source Python package for performing canonical correlation analysis (CCA). CCA is a multivariate analysis method for identifying relationships between sets of variables. Pyrcca supports CCA with or without regularization, and with or without linear, polynomial, or Gaussian kernelization. We first use an abstract example to describe Pyrcca functionality. We then demonstrate how Pyrcca can be used to analyze neuroimaging data. Specifically, we use Pyrcca to implement cross-subject comparison in a natural movie functional magnetic resonance imaging (fMRI) experiment by finding a data-driven set of functional response patterns that are similar across individuals. We validate this cross-subject comparison method in Pyrcca by predicting responses to novel natural movies across subjects. Finally, we show how Pyrcca can reveal retinotopic organization in brain responses to natural movies without the need for an explicit model.
Learning with Generalization Capability by Kernel Methods of Bounded Complexity
Czech Academy of Sciences Publication Activity Database
Kůrková, Věra; Sanguineti, M.
2005-01-01
Roč. 21, č. 3 (2005), s. 350-367 ISSN 0885-064X R&D Projects: GA AV ČR 1ET100300419 Institutional research plan: CEZ:AV0Z10300504 Keywords : supervised learning * generalization * model complexity * kernel methods * minimization of regularized empirical errors * upper bounds on rates of approximate optimization Subject RIV: BA - General Mathematics Impact factor: 1.186, year: 2005
On methods to increase the security of the Linux kernel
International Nuclear Information System (INIS)
Matvejchikov, I.V.
2014-01-01
Methods to increase the security of the Linux kernel for the implementation of imposed protection tools have been examined. The methods of incorporation into various subsystems of the kernel on the x86 architecture have been described [ru
A Novel Kernel-Based Regularization Technique for PET Image Reconstruction
Directory of Open Access Journals (Sweden)
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.
Directory of Open Access Journals (Sweden)
T. Borsdorff
2014-02-01
Full Text Available Insights are given into Tikhonov regularization and its application to the retrieval of vertical column densities of atmospheric trace gases from remote sensing measurements. The study builds upon the equivalence of the least-squares profile-scaling approach and Tikhonov regularization method of the first kind with an infinite regularization strength. Here, the vertical profile is expressed relative to a reference profile. On the basis of this, we propose a new algorithm as an extension of the least-squares profile scaling which permits the calculation of total column averaging kernels on arbitrary vertical grids using an analytic expression. Moreover, we discuss the effective null space of the retrieval, which comprises those parts of a vertical trace gas distribution which cannot be inferred from the measurements. Numerically the algorithm can be implemented in a robust and efficient manner. In particular for operational data processing with challenging demands on processing time, the proposed inversion method in combination with highly efficient forward models is an asset. For demonstration purposes, we apply the algorithm to CO column retrieval from simulated measurements in the 2.3 μm spectral region and to O3 column retrieval from the UV. These represent ideal measurements of a series of spaceborne spectrometers such as SCIAMACHY, TROPOMI, GOME, and GOME-2. For both spectral ranges, we consider clear-sky and cloudy scenes where clouds are modelled as an elevated Lambertian surface. Here, the smoothing error for the clear-sky and cloudy atmosphere is significant and reaches several percent, depending on the reference profile which is used for scaling. This underlines the importance of the column averaging kernel for a proper interpretation of retrieved column densities. Furthermore, we show that the smoothing due to regularization can be underestimated by calculating the column averaging kernel on a too coarse vertical grid. For both
Comparative Analysis of Kernel Methods for Statistical Shape Learning
National Research Council Canada - National Science Library
Rathi, Yogesh; Dambreville, Samuel; Tannenbaum, Allen
2006-01-01
.... In this work, we perform a comparative analysis of shape learning techniques such as linear PCA, kernel PCA, locally linear embedding and propose a new method, kernelized locally linear embedding...
Energy Technology Data Exchange (ETDEWEB)
Hao, Ming; Wang, Yanli, E-mail: ywang@ncbi.nlm.nih.gov; Bryant, Stephen H., E-mail: bryant@ncbi.nlm.nih.gov
2016-02-25
Identification of drug-target interactions (DTI) is a central task in drug discovery processes. In this work, a simple but effective regularized least squares integrating with nonlinear kernel fusion (RLS-KF) algorithm is proposed to perform DTI predictions. Using benchmark DTI datasets, our proposed algorithm achieves the state-of-the-art results with area under precision–recall curve (AUPR) of 0.915, 0.925, 0.853 and 0.909 for enzymes, ion channels (IC), G protein-coupled receptors (GPCR) and nuclear receptors (NR) based on 10 fold cross-validation. The performance can further be improved by using a recalculated kernel matrix, especially for the small set of nuclear receptors with AUPR of 0.945. Importantly, most of the top ranked interaction predictions can be validated by experimental data reported in the literature, bioassay results in the PubChem BioAssay database, as well as other previous studies. Our analysis suggests that the proposed RLS-KF is helpful for studying DTI, drug repositioning as well as polypharmacology, and may help to accelerate drug discovery by identifying novel drug targets. - Graphical abstract: Flowchart of the proposed RLS-KF algorithm for drug-target interaction predictions. - Highlights: • A nonlinear kernel fusion algorithm is proposed to perform drug-target interaction predictions. • Performance can further be improved by using the recalculated kernel. • Top predictions can be validated by experimental data.
International Nuclear Information System (INIS)
Hao, Ming; Wang, Yanli; Bryant, Stephen H.
2016-01-01
Identification of drug-target interactions (DTI) is a central task in drug discovery processes. In this work, a simple but effective regularized least squares integrating with nonlinear kernel fusion (RLS-KF) algorithm is proposed to perform DTI predictions. Using benchmark DTI datasets, our proposed algorithm achieves the state-of-the-art results with area under precision–recall curve (AUPR) of 0.915, 0.925, 0.853 and 0.909 for enzymes, ion channels (IC), G protein-coupled receptors (GPCR) and nuclear receptors (NR) based on 10 fold cross-validation. The performance can further be improved by using a recalculated kernel matrix, especially for the small set of nuclear receptors with AUPR of 0.945. Importantly, most of the top ranked interaction predictions can be validated by experimental data reported in the literature, bioassay results in the PubChem BioAssay database, as well as other previous studies. Our analysis suggests that the proposed RLS-KF is helpful for studying DTI, drug repositioning as well as polypharmacology, and may help to accelerate drug discovery by identifying novel drug targets. - Graphical abstract: Flowchart of the proposed RLS-KF algorithm for drug-target interaction predictions. - Highlights: • A nonlinear kernel fusion algorithm is proposed to perform drug-target interaction predictions. • Performance can further be improved by using the recalculated kernel. • Top predictions can be validated by experimental data.
Single pass kernel k-means clustering method
Indian Academy of Sciences (India)
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.
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
Regularization methods in Banach spaces
Schuster, Thomas; Hofmann, Bernd; Kazimierski, Kamil S
2012-01-01
Regularization methods aimed at finding stable approximate solutions are a necessary tool to tackle inverse and ill-posed problems. Usually the mathematical model of an inverse problem consists of an operator equation of the first kind and often the associated forward operator acts between Hilbert spaces. However, for numerous problems the reasons for using a Hilbert space setting seem to be based rather on conventions than on an approprimate and realistic model choice, so often a Banach space setting would be closer to reality. Furthermore, sparsity constraints using general Lp-norms or the B
Kernel based eigenvalue-decomposition methods for analysing ham
DEFF Research Database (Denmark)
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...
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.
Anatomically-aided PET reconstruction using the kernel method.
Hutchcroft, Will; Wang, Guobao; Chen, Kevin T; Catana, Ciprian; Qi, Jinyi
2016-09-21
This paper extends the kernel method that was proposed previously for dynamic PET reconstruction, to incorporate anatomical side information into the PET reconstruction model. In contrast to existing methods that incorporate anatomical information using a penalized likelihood framework, the proposed method incorporates this information in the simpler maximum likelihood (ML) formulation and is amenable to ordered subsets. The new method also does not require any segmentation of the anatomical image to obtain edge information. We compare the kernel method with the Bowsher method for anatomically-aided PET image reconstruction through a simulated data set. Computer simulations demonstrate that the kernel method offers advantages over the Bowsher method in region of interest quantification. Additionally the kernel method is applied to a 3D patient data set. The kernel method results in reduced noise at a matched contrast level compared with the conventional ML expectation maximization algorithm.
Dynamic PET Image reconstruction for parametric imaging using the HYPR kernel method
Spencer, Benjamin; Qi, Jinyi; Badawi, Ramsey D.; Wang, Guobao
2017-03-01
Dynamic PET image reconstruction is a challenging problem because of the ill-conditioned nature of PET and the lowcounting statistics resulted from short time-frames in dynamic imaging. The kernel method for image reconstruction has been developed to improve image reconstruction of low-count PET data by incorporating prior information derived from high-count composite data. In contrast to most of the existing regularization-based methods, the kernel method embeds image prior information in the forward projection model and does not require an explicit regularization term in the reconstruction formula. Inspired by the existing highly constrained back-projection (HYPR) algorithm for dynamic PET image denoising, we propose in this work a new type of kernel that is simpler to implement and further improves the kernel-based dynamic PET image reconstruction. Our evaluation study using a physical phantom scan with synthetic FDG tracer kinetics has demonstrated that the new HYPR kernel-based reconstruction can achieve a better region-of-interest (ROI) bias versus standard deviation trade-off for dynamic PET parametric imaging than the post-reconstruction HYPR denoising method and the previously used nonlocal-means kernel.
PKI, Gamma Radiation Reactor Shielding Calculation by Point-Kernel Method
International Nuclear Information System (INIS)
Li Chunhuai; Zhang Liwu; Zhang Yuqin; Zhang Chuanxu; Niu Xihua
1990-01-01
1 - Description of program or function: This code calculates radiation shielding problem of gamma-ray in geometric space. 2 - Method of solution: PKI uses a point kernel integration technique, describes radiation shielding geometric space by using geometric space configuration method and coordinate conversion, and makes use of calculation result of reactor primary shielding and flow regularity in loop system for coolant
Comparison of Kernel Equating and Item Response Theory Equating Methods
Meng, Yu
2012-01-01
The kernel method of test equating is a unified approach to test equating with some advantages over traditional equating methods. Therefore, it is important to evaluate in a comprehensive way the usefulness and appropriateness of the Kernel equating (KE) method, as well as its advantages and disadvantages compared with several popular item…
Kernel Methods for Machine Learning with Life Science Applications
DEFF Research Database (Denmark)
Abrahamsen, Trine Julie
Kernel methods refer to a family of widely used nonlinear algorithms for machine learning tasks like classification, regression, and feature extraction. By exploiting the so-called kernel trick straightforward extensions of classical linear algorithms are enabled as long as the data only appear a...
On the equivalence of different regularization methods
International Nuclear Information System (INIS)
Brzezowski, S.
1985-01-01
The R-circunflex-operation preceded by the regularization procedure is discussed. Some arguments are given, according to which the results may depend on the method of regularization, introduced in order to avoid divergences in perturbation calculations. 10 refs. (author)
Deep kernel learning method for SAR image target recognition
Chen, Xiuyuan; Peng, Xiyuan; Duan, Ran; Li, Junbao
2017-10-01
With the development of deep learning, research on image target recognition has made great progress in recent years. Remote sensing detection urgently requires target recognition for military, geographic, and other scientific research. This paper aims to solve the synthetic aperture radar image target recognition problem by combining deep and kernel learning. The model, which has a multilayer multiple kernel structure, is optimized layer by layer with the parameters of Support Vector Machine and a gradient descent algorithm. This new deep kernel learning method improves accuracy and achieves competitive recognition results compared with other learning methods.
Application of Turchin's method of statistical regularization
Zelenyi, Mikhail; Poliakova, Mariia; Nozik, Alexander; Khudyakov, Alexey
2018-04-01
During analysis of experimental data, one usually needs to restore a signal after it has been convoluted with some kind of apparatus function. According to Hadamard's definition this problem is ill-posed and requires regularization to provide sensible results. In this article we describe an implementation of the Turchin's method of statistical regularization based on the Bayesian approach to the regularization strategy.
Improved modeling of clinical data with kernel methods.
Daemen, Anneleen; Timmerman, Dirk; Van den Bosch, Thierry; Bottomley, Cecilia; Kirk, Emma; Van Holsbeke, Caroline; Valentin, Lil; Bourne, Tom; De Moor, Bart
2012-02-01
Despite the rise of high-throughput technologies, clinical data such as age, gender and medical history guide clinical management for most diseases and examinations. To improve clinical management, available patient information should be fully exploited. This requires appropriate modeling of relevant parameters. When kernel methods are used, traditional kernel functions such as the linear kernel are often applied to the set of clinical parameters. These kernel functions, however, have their disadvantages due to the specific characteristics of clinical data, being a mix of variable types with each variable its own range. We propose a new kernel function specifically adapted to the characteristics of clinical data. The clinical kernel function provides a better representation of patients' similarity by equalizing the influence of all variables and taking into account the range r of the variables. Moreover, it is robust with respect to changes in r. Incorporated in a least squares support vector machine, the new kernel function results in significantly improved diagnosis, prognosis and prediction of therapy response. This is illustrated on four clinical data sets within gynecology, with an average increase in test area under the ROC curve (AUC) of 0.023, 0.021, 0.122 and 0.019, respectively. Moreover, when combining clinical parameters and expression data in three case studies on breast cancer, results improved overall with use of the new kernel function and when considering both data types in a weighted fashion, with a larger weight assigned to the clinical parameters. The increase in AUC with respect to a standard kernel function and/or unweighted data combination was maximum 0.127, 0.042 and 0.118 for the three case studies. For clinical data consisting of variables of different types, the proposed kernel function--which takes into account the type and range of each variable--has shown to be a better alternative for linear and non-linear classification problems
DEFF Research Database (Denmark)
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...
Iterative Regularization with Minimum-Residual Methods
DEFF Research Database (Denmark)
Jensen, Toke Koldborg; Hansen, Per Christian
2007-01-01
subspaces. We provide a combination of theory and numerical examples, and our analysis confirms the experience that MINRES and MR-II can work as general regularization methods. We also demonstrate theoretically and experimentally that the same is not true, in general, for GMRES and RRGMRES their success......We study the regularization properties of iterative minimum-residual methods applied to discrete ill-posed problems. In these methods, the projection onto the underlying Krylov subspace acts as a regularizer, and the emphasis of this work is on the role played by the basis vectors of these Krylov...... as regularization methods is highly problem dependent....
Iterative regularization with minimum-residual methods
DEFF Research Database (Denmark)
Jensen, Toke Koldborg; Hansen, Per Christian
2006-01-01
subspaces. We provide a combination of theory and numerical examples, and our analysis confirms the experience that MINRES and MR-II can work as general regularization methods. We also demonstrate theoretically and experimentally that the same is not true, in general, for GMRES and RRGMRES - their success......We study the regularization properties of iterative minimum-residual methods applied to discrete ill-posed problems. In these methods, the projection onto the underlying Krylov subspace acts as a regularizer, and the emphasis of this work is on the role played by the basis vectors of these Krylov...... as regularization methods is highly problem dependent....
Analysis and regularization of the thin-wire integral equation with reduced kernel
Beurden, van M.C.; Tijhuis, A.G.
2007-01-01
For the straight wire, modeled as a hollow tube, we establish a conditional equivalence relation between the integral equations with exact and reduced kernel. This relation allows us to examine the existence and uniqueness conditions for the integral equation with reduced kernel, based on a local
Input Space Regularization Stabilizes Pre-images for Kernel PCA De-noising
DEFF Research Database (Denmark)
Abrahamsen, Trine Julie; Hansen, Lars Kai
2009-01-01
Solution of the pre-image problem is key to efficient nonlinear de-noising using kernel Principal Component Analysis. Pre-image estimation is inherently ill-posed for typical kernels used in applications and consequently the most widely used estimation schemes lack stability. For de...
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.
Local coding based matching kernel method for image classification.
Directory of Open Access Journals (Sweden)
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.
A method for manufacturing kernels of metallic oxides and the thus obtained kernels
International Nuclear Information System (INIS)
Lelievre Bernard; Feugier, Andre.
1973-01-01
A method is described for manufacturing fissile or fertile metal oxide kernels, consisting in adding at least a chemical compound capable of releasing ammonia to an aqueous solution of actinide nitrates dispersing the thus obtained solution dropwise in a hot organic phase so as to gelify the drops and transform them into solid particles, washing drying and treating said particles so as to transform them into oxide kernels. Such a method is characterized in that the organic phase used in the gel-forming reactions comprises a mixture of two organic liquids, one of which acts as a solvent, whereas the other is a product capable of extracting the metal-salt anions from the drops while the gel forming reaction is taking place. This can be applied to the so-called high temperature nuclear reactors [fr
Anatomical image-guided fluorescence molecular tomography reconstruction using kernel method
Baikejiang, Reheman; Zhao, Yue; Fite, Brett Z.; Ferrara, Katherine W.; Li, Changqing
2017-01-01
Abstract. Fluorescence molecular tomography (FMT) is an important in vivo imaging modality to visualize physiological and pathological processes in small animals. However, FMT reconstruction is ill-posed and ill-conditioned due to strong optical scattering in deep tissues, which results in poor spatial resolution. It is well known that FMT image quality can be improved substantially by applying the structural guidance in the FMT reconstruction. An approach to introducing anatomical information into the FMT reconstruction is presented using the kernel method. In contrast to conventional methods that incorporate anatomical information with a Laplacian-type regularization matrix, the proposed method introduces the anatomical guidance into the projection model of FMT. The primary advantage of the proposed method is that it does not require segmentation of targets in the anatomical images. Numerical simulations and phantom experiments have been performed to demonstrate the proposed approach’s feasibility. Numerical simulation results indicate that the proposed kernel method can separate two FMT targets with an edge-to-edge distance of 1 mm and is robust to false-positive guidance and inhomogeneity in the anatomical image. For the phantom experiments with two FMT targets, the kernel method has reconstructed both targets successfully, which further validates the proposed kernel method. PMID:28464120
Reproducing Kernel Method for Solving Nonlinear Differential-Difference Equations
Directory of Open Access Journals (Sweden)
Reza Mokhtari
2012-01-01
Full Text Available On the basis of reproducing kernel Hilbert spaces theory, an iterative algorithm for solving some nonlinear differential-difference equations (NDDEs is presented. The analytical solution is shown in a series form in a reproducing kernel space, and the approximate solution , is constructed by truncating the series to terms. The convergence of , to the analytical solution is also proved. Results obtained by the proposed method imply that it can be considered as a simple and accurate method for solving such differential-difference problems.
Employment of kernel methods on wind turbine power performance assessment
DEFF Research Database (Denmark)
Skrimpas, Georgios Alexandros; Sweeney, Christian Walsted; Marhadi, Kun S.
2015-01-01
A power performance assessment technique is developed for the detection of power production discrepancies in wind turbines. The method employs a widely used nonparametric pattern recognition technique, the kernel methods. The evaluation is based on the trending of an extracted feature from...... the kernel matrix, called similarity index, which is introduced by the authors for the first time. The operation of the turbine and consequently the computation of the similarity indexes is classified into five power bins offering better resolution and thus more consistent root cause analysis. The accurate...
Single pass kernel k-means clustering method
Indian Academy of Sciences (India)
In unsupervised classiﬁcation, kernel -means clustering method has been shown to perform better than conventional -means clustering method in ... 518501, India; Department of Computer Science and Engineering, Jawaharlal Nehru Technological University, Anantapur College of Engineering, Anantapur 515002, India ...
Directory of Open Access Journals (Sweden)
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.
Kernel Methods for Mining Instance Data in Ontologies
Bloehdorn, Stephan; Sure, York
The amount of ontologies and meta data available on the Web is constantly growing. The successful application of machine learning techniques for learning of ontologies from textual data, i.e. mining for the Semantic Web, contributes to this trend. However, no principal approaches exist so far for mining from the Semantic Web. We investigate how machine learning algorithms can be made amenable for directly taking advantage of the rich knowledge expressed in ontologies and associated instance data. Kernel methods have been successfully employed in various learning tasks and provide a clean framework for interfacing between non-vectorial data and machine learning algorithms. In this spirit, we express the problem of mining instances in ontologies as the problem of defining valid corresponding kernels. We present a principled framework for designing such kernels by means of decomposing the kernel computation into specialized kernels for selected characteristics of an ontology which can be flexibly assembled and tuned. Initial experiments on real world Semantic Web data enjoy promising results and show the usefulness of our approach.
Diagrammatic methods in phase-space regularization
International Nuclear Information System (INIS)
Bern, Z.; Halpern, M.B.; California Univ., Berkeley
1987-11-01
Using the scalar prototype and gauge theory as the simplest possible examples, diagrammatic methods are developed for the recently proposed phase-space form of continuum regularization. A number of one-loop and all-order applications are given, including general diagrammatic discussions of the nogrowth theorem and the uniqueness of the phase-space stochastic calculus. The approach also generates an alternate derivation of the equivalence of the large-β phase-space regularization to the more conventional coordinate-space regularization. (orig.)
Regularized Pre-image Estimation for Kernel PCA De-noising
DEFF Research Database (Denmark)
Abrahamsen, Trine Julie; Hansen, Lars Kai
2011-01-01
The main challenge in de-noising by kernel Principal Component Analysis (PCA) is the mapping of de-noised feature space points back into input space, also referred to as “the pre-image problem”. Since the feature space mapping is typically not bijective, pre-image estimation is inherently illposed...
Jacquin, Laval; Cao, Tuong-Vi; Ahmadi, Nourollah
2016-01-01
One objective of this study was to provide readers with a clear and unified understanding of parametric statistical and kernel methods, used for genomic prediction, and to compare some of these in the context of rice breeding for quantitative traits. Furthermore, another objective was to provide a simple and user-friendly R package, named KRMM, which allows users to perform RKHS regression with several kernels. After introducing the concept of regularized empirical risk minimization, the connections between well-known parametric and kernel methods such as Ridge regression [i.e., genomic best linear unbiased predictor (GBLUP)] and reproducing kernel Hilbert space (RKHS) regression were reviewed. Ridge regression was then reformulated so as to show and emphasize the advantage of the kernel "trick" concept, exploited by kernel methods in the context of epistatic genetic architectures, over parametric frameworks used by conventional methods. Some parametric and kernel methods; least absolute shrinkage and selection operator (LASSO), GBLUP, support vector machine regression (SVR) and RKHS regression were thereupon compared for their genomic predictive ability in the context of rice breeding using three real data sets. Among the compared methods, RKHS regression and SVR were often the most accurate methods for prediction followed by GBLUP and LASSO. An R function which allows users to perform RR-BLUP of marker effects, GBLUP and RKHS regression, with a Gaussian, Laplacian, polynomial or ANOVA kernel, in a reasonable computation time has been developed. Moreover, a modified version of this function, which allows users to tune kernels for RKHS regression, has also been developed and parallelized for HPC Linux clusters. The corresponding KRMM package and all scripts have been made publicly available.
Kernel based pattern analysis methods using eigen-decompositions for reading Icelandic sagas
DEFF Research Database (Denmark)
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....
Aveiro method in reproducing kernel Hilbert spaces under complete dictionary
Mai, Weixiong; Qian, Tao
2017-12-01
Aveiro Method is a sparse representation method in reproducing kernel Hilbert spaces (RKHS) that gives orthogonal projections in linear combinations of reproducing kernels over uniqueness sets. It, however, suffers from determination of uniqueness sets in the underlying RKHS. In fact, in general spaces, uniqueness sets are not easy to be identified, let alone the convergence speed aspect with Aveiro Method. To avoid those difficulties we propose an anew Aveiro Method based on a dictionary and the matching pursuit idea. What we do, in fact, are more: The new Aveiro method will be in relation to the recently proposed, the so called Pre-Orthogonal Greedy Algorithm (P-OGA) involving completion of a given dictionary. The new method is called Aveiro Method Under Complete Dictionary (AMUCD). The complete dictionary consists of all directional derivatives of the underlying reproducing kernels. We show that, under the boundary vanishing condition, bring available for the classical Hardy and Paley-Wiener spaces, the complete dictionary enables an efficient expansion of any given element in the Hilbert space. The proposed method reveals new and advanced aspects in both the Aveiro Method and the greedy algorithm.
Machine learning properties of materials and molecules with entropy-regularized kernels
Ceriotti, Michele; Bartók, Albert; CsáNyi, GáBor; de, Sandip
Application of machine-learning methods to physics, chemistry and materials science is gaining traction as a strategy to obtain accurate predictions of the properties of matter at a fraction of the typical cost of quantum mechanical electronic structure calculations. In this endeavor, one can leverage general-purpose frameworks for supervised-learning. It is however very important that the input data - for instance the positions of atoms in a molecule or solid - is processed into a form that reflects all the underlying physical symmetries of the problem, and that possesses the regularity properties that are required by machine-learning algorithms. Here we introduce a general strategy to build a representation of this kind. We will start from existing approaches to compare local environments (basically, groups of atoms), and combine them using techniques borrowed from optimal transport theory, discussing the relation between this idea and additive energy decompositions. We will present a few examples demonstrating the potential of this approach as a tool to predict molecular and materials' properties with an accuracy on par with state-of-the-art electronic structure methods. MARVEL NCCR (Swiss National Science Foundation) and ERC StG HBMAP (European Research Council, G.A. 677013).
A Fast Multiple-Kernel Method With Applications to Detect Gene-Environment Interaction.
Marceau, Rachel; Lu, Wenbin; Holloway, Shannon; Sale, Michèle M; Worrall, Bradford B; Williams, Stephen R; Hsu, Fang-Chi; Tzeng, Jung-Ying
2015-09-01
Kernel machine (KM) models are a powerful tool for exploring associations between sets of genetic variants and complex traits. Although most KM methods use a single kernel function to assess the marginal effect of a variable set, KM analyses involving multiple kernels have become increasingly popular. Multikernel analysis allows researchers to study more complex problems, such as assessing gene-gene or gene-environment interactions, incorporating variance-component based methods for population substructure into rare-variant association testing, and assessing the conditional effects of a variable set adjusting for other variable sets. The KM framework is robust, powerful, and provides efficient dimension reduction for multifactor analyses, but requires the estimation of high dimensional nuisance parameters. Traditional estimation techniques, including regularization and the "expectation-maximization (EM)" algorithm, have a large computational cost and are not scalable to large sample sizes needed for rare variant analysis. Therefore, under the context of gene-environment interaction, we propose a computationally efficient and statistically rigorous "fastKM" algorithm for multikernel analysis that is based on a low-rank approximation to the nuisance effect kernel matrices. Our algorithm is applicable to various trait types (e.g., continuous, binary, and survival traits) and can be implemented using any existing single-kernel analysis software. Through extensive simulation studies, we show that our algorithm has similar performance to an EM-based KM approach for quantitative traits while running much faster. We also apply our method to the Vitamin Intervention for Stroke Prevention (VISP) clinical trial, examining gene-by-vitamin effects on recurrent stroke risk and gene-by-age effects on change in homocysteine level. © 2015 WILEY PERIODICALS, INC.
A kernel plus method for quantifying wind turbine performance upgrades
Lee, Giwhyun
2014-04-21
Power curves are commonly estimated using the binning method recommended by the International Electrotechnical Commission, which primarily incorporates wind speed information. When such power curves are used to quantify a turbine\\'s upgrade, the results may not be accurate because many other environmental factors in addition to wind speed, such as temperature, air pressure, turbulence intensity, wind shear and humidity, all potentially affect the turbine\\'s power output. Wind industry practitioners are aware of the need to filter out effects from environmental conditions. Toward that objective, we developed a kernel plus method that allows incorporation of multivariate environmental factors in a power curve model, thereby controlling the effects from environmental factors while comparing power outputs. We demonstrate that the kernel plus method can serve as a useful tool for quantifying a turbine\\'s upgrade because it is sensitive to small and moderate changes caused by certain turbine upgrades. Although we demonstrate the utility of the kernel plus method in this specific application, the resulting method is a general, multivariate model that can connect other physical factors, as long as their measurements are available, with a turbine\\'s power output, which may allow us to explore new physical properties associated with wind turbine performance. © 2014 John Wiley & Sons, Ltd.
Enriched reproducing kernel particle method for fractional advection-diffusion equation
Ying, Yuping; Lian, Yanping; Tang, Shaoqiang; Liu, Wing Kam
2018-06-01
The reproducing kernel particle method (RKPM) has been efficiently applied to problems with large deformations, high gradients and high modal density. In this paper, it is extended to solve a nonlocal problem modeled by a fractional advection-diffusion equation (FADE), which exhibits a boundary layer with low regularity. We formulate this method on a moving least-square approach. Via the enrichment of fractional-order power functions to the traditional integer-order basis for RKPM, leading terms of the solution to the FADE can be exactly reproduced, which guarantees a good approximation to the boundary layer. Numerical tests are performed to verify the proposed approach.
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.
Directory of Open Access Journals (Sweden)
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.
Wang, Jim Jing-Yan; Huang, Jianhua Z.; Sun, Yijun; Gao, Xin
2014-01-01
by regularizing NMF with a nearest neighbor graph constructed from the input data set. However, GNMF has two main bottlenecks. First, using the original feature space directly to construct the graph is not necessarily optimal because of the noisy and irrelevant
Kernel methods for large-scale genomic data analysis
Xing, Eric P.; Schaid, Daniel J.
2015-01-01
Machine learning, particularly kernel methods, has been demonstrated as a promising new tool to tackle the challenges imposed by today’s explosive data growth in genomics. They provide a practical and principled approach to learning how a large number of genetic variants are associated with complex phenotypes, to help reveal the complexity in the relationship between the genetic markers and the outcome of interest. In this review, we highlight the potential key role it will have in modern genomic data processing, especially with regard to integration with classical methods for gene prioritizing, prediction and data fusion. PMID:25053743
Kernel method for clustering based on optimal target vector
International Nuclear Information System (INIS)
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
Linear and kernel methods for multi- and hypervariate change detection
DEFF Research Database (Denmark)
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...
Kernel methods in orthogonalization of multi- and hypervariate data
DEFF Research Database (Denmark)
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...
A high-order SPH method by introducing inverse kernels
Directory of Open Access Journals (Sweden)
Le Fang
2017-02-01
Full Text Available The smoothed particle hydrodynamics (SPH method is usually expected to be an efficient numerical tool for calculating the fluid-structure interactions in compressors; however, an endogenetic restriction is the problem of low-order consistency. A high-order SPH method by introducing inverse kernels, which is quite easy to be implemented but efficient, is proposed for solving this restriction. The basic inverse method and the special treatment near boundary are introduced with also the discussion of the combination of the Least-Square (LS and Moving-Least-Square (MLS methods. Then detailed analysis in spectral space is presented for people to better understand this method. Finally we show three test examples to verify the method behavior.
A laser optical method for detecting corn kernel defects
Energy Technology Data Exchange (ETDEWEB)
Gunasekaran, S.; Paulsen, M. R.; Shove, G. C.
1984-01-01
An opto-electronic instrument was developed to examine individual corn kernels and detect various kernel defects according to reflectance differences. A low power helium-neon (He-Ne) laser (632.8 nm, red light) was used as the light source in the instrument. Reflectance from good and defective parts of corn kernel surfaces differed by approximately 40%. Broken, chipped, and starch-cracked kernels were detected with nearly 100% accuracy; while surface-split kernels were detected with about 80% accuracy. (author)
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
International Nuclear Information System (INIS)
Wang, Wenyan; Han, Bo; Yamamoto, Masahiro
2013-01-01
We propose a new numerical method for reproducing kernel Hilbert space to solve an inverse source problem for a two-dimensional fractional diffusion equation, where we are required to determine an x-dependent function in a source term by data at the final time. The exact solution is represented in the form of a series and the approximation solution is obtained by truncating the series. Furthermore, a technique is proposed to improve some of the existing methods. We prove that the numerical method is convergent under an a priori assumption of the regularity of solutions. The method is simple to implement. Our numerical result shows that our method is effective and that it is robust against noise in L 2 -space in reconstructing a source function. (paper)
An iterative method for Tikhonov regularization with a general linear regularization operator
Hochstenbach, M.E.; Reichel, L.
2010-01-01
Tikhonov regularization is one of the most popular approaches to solve discrete ill-posed problems with error-contaminated data. A regularization operator and a suitable value of a regularization parameter have to be chosen. This paper describes an iterative method, based on Golub-Kahan
Kernel based methods for accelerated failure time model with ultra-high dimensional data
Directory of Open Access Journals (Sweden)
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.
Directory of Open Access Journals (Sweden)
Laval Jacquin
2016-08-01
Full Text Available One objective of this study was to provide readers with a clear and unified understanding ofparametric statistical and kernel methods, used for genomic prediction, and to compare some ofthese in the context of rice breeding for quantitative traits. Furthermore, another objective wasto provide a simple and user-friendly R package, named KRMM, which allows users to performRKHS regression with several kernels. After introducing the concept of regularized empiricalrisk minimization, the connections between well-known parametric and kernel methods suchas Ridge regression (i.e. genomic best linear unbiased predictor (GBLUP and reproducingkernel Hilbert space (RKHS regression were reviewed. Ridge regression was then reformulatedso as to show and emphasize the advantage of the kernel trick concept, exploited by kernelmethods in the context of epistatic genetic architectures, over parametric frameworks used byconventional methods. Some parametric and kernel methods; least absolute shrinkage andselection operator (LASSO, GBLUP, support vector machine regression (SVR and RKHSregression were thereupon compared for their genomic predictive ability in the context of ricebreeding using three real data sets. Among the compared methods, RKHS regression and SVRwere often the most accurate methods for prediction followed by GBLUP and LASSO. An Rfunction which allows users to perform RR-BLUP of marker effects, GBLUP and RKHS regression,with a Gaussian, Laplacian, polynomial or ANOVA kernel, in a reasonable computation time hasbeen developed. Moreover, a modified version of this function, which allows users to tune kernelsfor RKHS regression, has also been developed and parallelized for HPC Linux clusters. The corresponding KRMM package and all scripts have been made publicly available.
Point kernels and superposition methods for scatter dose calculations in brachytherapy
International Nuclear Information System (INIS)
Carlsson, A.K.
2000-01-01
Point kernels have been generated and applied for calculation of scatter dose distributions around monoenergetic point sources for photon energies ranging from 28 to 662 keV. Three different approaches for dose calculations have been compared: a single-kernel superposition method, a single-kernel superposition method where the point kernels are approximated as isotropic and a novel 'successive-scattering' superposition method for improved modelling of the dose from multiply scattered photons. An extended version of the EGS4 Monte Carlo code was used for generating the kernels and for benchmarking the absorbed dose distributions calculated with the superposition methods. It is shown that dose calculation by superposition at and below 100 keV can be simplified by using isotropic point kernels. Compared to the assumption of full in-scattering made by algorithms currently in clinical use, the single-kernel superposition method improves dose calculations in a half-phantom consisting of air and water. Further improvements are obtained using the successive-scattering superposition method, which reduces the overestimates of dose close to the phantom surface usually associated with kernel superposition methods at brachytherapy photon energies. It is also shown that scatter dose point kernels can be parametrized to biexponential functions, making them suitable for use with an effective implementation of the collapsed cone superposition algorithm. (author)
Open Problem: Kernel methods on manifolds and metric spaces
DEFF Research Database (Denmark)
Feragen, Aasa; Hauberg, Søren
2016-01-01
Radial kernels are well-suited for machine learning over general geodesic metric spaces, where pairwise distances are often the only computable quantity available. We have recently shown that geodesic exponential kernels are only positive definite for all bandwidths when the input space has strong...... linear properties. This negative result hints that radial kernel are perhaps not suitable over geodesic metric spaces after all. Here, however, we present evidence that large intervals of bandwidths exist where geodesic exponential kernels have high probability of being positive definite over finite...... datasets, while still having significant predictive power. From this we formulate conjectures on the probability of a positive definite kernel matrix for a finite random sample, depending on the geometry of the data space and the spread of the sample....
Application of the kernel method to the inverse geosounding problem.
Hidalgo, Hugo; Sosa León, Sonia; Gómez-Treviño, Enrique
2003-01-01
Determining the layered structure of the earth demands the solution of a variety of inverse problems; in the case of electromagnetic soundings at low induction numbers, the problem is linear, for the measurements may be represented as a linear functional of the electrical conductivity distribution. In this paper, an application of the support vector (SV) regression technique to the inversion of electromagnetic data is presented. We take advantage of the regularizing properties of the SV learning algorithm and use it as a modeling technique with synthetic and field data. The SV method presents better recovery of synthetic models than Tikhonov's regularization. As the SV formulation is solved in the space of the data, which has a small dimension in this application, a smaller problem than that considered with Tikhonov's regularization is produced. For field data, the SV formulation develops models similar to those obtained via linear programming techniques, but with the added characteristic of robustness.
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
Linear and kernel methods for multivariate change detection
DEFF Research Database (Denmark)
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...
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
International Nuclear Information System (INIS)
Lomnitz-Adler, J.; Brink, D.M.
1976-01-01
A generating function for the eigenvalues of the RGM Normalization Kernel is expressed in terms of the diagonal matrix elements of thw GCM Overlap Kernel. An asymptotic expression for the eigenvalues is obtained by using the Method of Steepest Descent. (Auth.)
Kernel methods for interpretable machine learning of order parameters
Ponte, Pedro; Melko, Roger G.
2017-11-01
Machine learning is capable of discriminating phases of matter, and finding associated phase transitions, directly from large data sets of raw state configurations. In the context of condensed matter physics, most progress in the field of supervised learning has come from employing neural networks as classifiers. Although very powerful, such algorithms suffer from a lack of interpretability, which is usually desired in scientific applications in order to associate learned features with physical phenomena. In this paper, we explore support vector machines (SVMs), which are a class of supervised kernel methods that provide interpretable decision functions. We find that SVMs can learn the mathematical form of physical discriminators, such as order parameters and Hamiltonian constraints, for a set of two-dimensional spin models: the ferromagnetic Ising model, a conserved-order-parameter Ising model, and the Ising gauge theory. The ability of SVMs to provide interpretable classification highlights their potential for automating feature detection in both synthetic and experimental data sets for condensed matter and other many-body systems.
Soft and hard classification by reproducing kernel Hilbert space methods.
Wahba, Grace
2002-12-24
Reproducing kernel Hilbert space (RKHS) methods provide a unified context for solving a wide variety of statistical modelling and function estimation problems. We consider two such problems: We are given a training set [yi, ti, i = 1, em leader, n], where yi is the response for the ith subject, and ti is a vector of attributes for this subject. The value of y(i) is a label that indicates which category it came from. For the first problem, we wish to build a model from the training set that assigns to each t in an attribute domain of interest an estimate of the probability pj(t) that a (future) subject with attribute vector t is in category j. The second problem is in some sense less ambitious; it is to build a model that assigns to each t a label, which classifies a future subject with that t into one of the categories or possibly "none of the above." The approach to the first of these two problems discussed here is a special case of what is known as penalized likelihood estimation. The approach to the second problem is known as the support vector machine. We also note some alternate but closely related approaches to the second problem. These approaches are all obtained as solutions to optimization problems in RKHS. Many other problems, in particular the solution of ill-posed inverse problems, can be obtained as solutions to optimization problems in RKHS and are mentioned in passing. We caution the reader that although a large literature exists in all of these topics, in this inaugural article we are selectively highlighting work of the author, former students, and other collaborators.
A blind deconvolution method based on L1/L2 regularization prior in the gradient space
Cai, Ying; Shi, Yu; Hua, Xia
2018-02-01
In the process of image restoration, the result of image restoration is very different from the real image because of the existence of noise, in order to solve the ill posed problem in image restoration, a blind deconvolution method based on L1/L2 regularization prior to gradient domain is proposed. The method presented in this paper first adds a function to the prior knowledge, which is the ratio of the L1 norm to the L2 norm, and takes the function as the penalty term in the high frequency domain of the image. Then, the function is iteratively updated, and the iterative shrinkage threshold algorithm is applied to solve the high frequency image. In this paper, it is considered that the information in the gradient domain is better for the estimation of blur kernel, so the blur kernel is estimated in the gradient domain. This problem can be quickly implemented in the frequency domain by fast Fast Fourier Transform. In addition, in order to improve the effectiveness of the algorithm, we have added a multi-scale iterative optimization method. This paper proposes the blind deconvolution method based on L1/L2 regularization priors in the gradient space can obtain the unique and stable solution in the process of image restoration, which not only keeps the edges and details of the image, but also ensures the accuracy of the results.
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.
Dose calculation methods in photon beam therapy using energy deposition kernels
International Nuclear Information System (INIS)
Ahnesjoe, A.
1991-01-01
The problem of calculating accurate dose distributions in treatment planning of megavoltage photon radiation therapy has been studied. New dose calculation algorithms using energy deposition kernels have been developed. The kernels describe the transfer of energy by secondary particles from a primary photon interaction site to its surroundings. Monte Carlo simulations of particle transport have been used for derivation of kernels for primary photon energies form 0.1 MeV to 50 MeV. The trade off between accuracy and calculational speed has been addressed by the development of two algorithms; one point oriented with low computional overhead for interactive use and one for fast and accurate calculation of dose distributions in a 3-dimensional lattice. The latter algorithm models secondary particle transport in heterogeneous tissue by scaling energy deposition kernels with the electron density of the tissue. The accuracy of the methods has been tested using full Monte Carlo simulations for different geometries, and found to be superior to conventional algorithms based on scaling of broad beam dose distributions. Methods have also been developed for characterization of clinical photon beams in entities appropriate for kernel based calculation models. By approximating the spectrum as laterally invariant, an effective spectrum and dose distribution for contaminating charge particles are derived form depth dose distributions measured in water, using analytical constraints. The spectrum is used to calculate kernels by superposition of monoenergetic kernels. The lateral energy fluence distribution is determined by deconvolving measured lateral dose distributions by a corresponding pencil beam kernel. Dose distributions for contaminating photons are described using two different methods, one for estimation of the dose outside of the collimated beam, and the other for calibration of output factors derived from kernel based dose calculations. (au)
Laplacian manifold regularization method for fluorescence molecular tomography
He, Xuelei; Wang, Xiaodong; Yi, Huangjian; Chen, Yanrong; Zhang, Xu; Yu, Jingjing; He, Xiaowei
2017-04-01
Sparse regularization methods have been widely used in fluorescence molecular tomography (FMT) for stable three-dimensional reconstruction. Generally, ℓ1-regularization-based methods allow for utilizing the sparsity nature of the target distribution. However, in addition to sparsity, the spatial structure information should be exploited as well. A joint ℓ1 and Laplacian manifold regularization model is proposed to improve the reconstruction performance, and two algorithms (with and without Barzilai-Borwein strategy) are presented to solve the regularization model. Numerical studies and in vivo experiment demonstrate that the proposed Gradient projection-resolved Laplacian manifold regularization method for the joint model performed better than the comparative algorithm for ℓ1 minimization method in both spatial aggregation and location accuracy.
Ambrogioni, Luca; Güçlü, Umut; van Gerven, Marcel A. J.; Maris, Eric
2017-01-01
This paper introduces the kernel mixture network, a new method for nonparametric estimation of conditional probability densities using neural networks. We model arbitrarily complex conditional densities as linear combinations of a family of kernel functions centered at a subset of training points. The weights are determined by the outer layer of a deep neural network, trained by minimizing the negative log likelihood. This generalizes the popular quantized softmax approach, which can be seen ...
Directory of Open Access Journals (Sweden)
Akgül Ali
2016-01-01
Full Text Available In this manuscript we investigate electrodynamic flow. For several values of the intimate parameters we proved that the approximate solution depends on a reproducing kernel model. Obtained results prove that the reproducing kernel method (RKM is very effective. We obtain good results without any transformation or discretization. Numerical experiments on test examples show that our proposed schemes are of high accuracy and strongly support the theoretical results.
International Nuclear Information System (INIS)
Matijevic, M.; Grgic, D.; Jecmenica, R.
2016-01-01
This paper presents comparison of the Krsko Power Plant simplified Spent Fuel Pool (SFP) dose rates using different computational shielding methodologies. The analysis was performed to estimate limiting gamma dose rates on wall mounted level instrumentation in case of significant loss of cooling water. The SFP was represented with simple homogenized cylinders (point kernel and Monte Carlo (MC)) or cuboids (MC) using uranium, iron, water, and dry-air as bulk region materials. The pool is divided on the old and new section where the old one has three additional subsections representing fuel assemblies (FAs) with different burnup/cooling time (60 days, 1 year and 5 years). The new section represents the FAs with the cooling time of 10 years. The time dependent fuel assembly isotopic composition was calculated using ORIGEN2 code applied to the depletion of one of the fuel assemblies present in the pool (AC-29). The source used in Microshield calculation is based on imported isotopic activities. The time dependent photon spectra with total source intensity from Microshield multigroup point kernel calculations was then prepared for two hybrid deterministic-stochastic sequences. One is based on SCALE/MAVRIC (Monaco and Denovo) methodology and another uses Monte Carlo code MCNP6.1.1b and ADVANTG3.0.1. code. Even though this model is a fairly simple one, the layers of shielding materials are thick enough to pose a significant shielding problem for MC method without the use of effective variance reduction (VR) technique. For that purpose the ADVANTG code was used to generate VR parameters (SB cards in SDEF and WWINP file) for MCNP fixed-source calculation using continuous energy transport. ADVATNG employs a deterministic forward-adjoint transport solver Denovo which implements CADIS/FW-CADIS methodology. Denovo implements a structured, Cartesian-grid SN solver based on the Koch-Baker-Alcouffe parallel transport sweep algorithm across x-y domain blocks. This was first
Energy Technology Data Exchange (ETDEWEB)
Sheu, R.-D.; Chui, C.-S.; Jiang, S.-H. E-mail: shjiang@mx.nthu.edu.tw
2003-12-01
A simplified method, based on the integral of the first collision kernel, is presented for performing gamma-ray skyshine calculations for the collimated sources. The first collision kernels were calculated in air for a reference air density by use of the EGS4 Monte Carlo code. These kernels can be applied to other air densities by applying density corrections. The integral first collision kernel (IFCK) method has been used to calculate two of the ANSI/ANS skyshine benchmark problems and the results were compared with a number of other commonly used codes. Our results were generally in good agreement with others but only spend a small fraction of the computation time required by the Monte Carlo calculations. The scheme of the IFCK method for dealing with lots of source collimation geometry is also presented in this study.
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.
A comprehensive benchmark of kernel methods to extract protein-protein interactions from literature.
Directory of Open Access Journals (Sweden)
Domonkos Tikk
Full Text Available The most important way of conveying new findings in biomedical research is scientific publication. Extraction of protein-protein interactions (PPIs reported in scientific publications is one of the core topics of text mining in the life sciences. Recently, a new class of such methods has been proposed - convolution kernels that identify PPIs using deep parses of sentences. However, comparing published results of different PPI extraction methods is impossible due to the use of different evaluation corpora, different evaluation metrics, different tuning procedures, etc. In this paper, we study whether the reported performance metrics are robust across different corpora and learning settings and whether the use of deep parsing actually leads to an increase in extraction quality. Our ultimate goal is to identify the one method that performs best in real-life scenarios, where information extraction is performed on unseen text and not on specifically prepared evaluation data. We performed a comprehensive benchmarking of nine different methods for PPI extraction that use convolution kernels on rich linguistic information. Methods were evaluated on five different public corpora using cross-validation, cross-learning, and cross-corpus evaluation. Our study confirms that kernels using dependency trees generally outperform kernels based on syntax trees. However, our study also shows that only the best kernel methods can compete with a simple rule-based approach when the evaluation prevents information leakage between training and test corpora. Our results further reveal that the F-score of many approaches drops significantly if no corpus-specific parameter optimization is applied and that methods reaching a good AUC score often perform much worse in terms of F-score. We conclude that for most kernels no sensible estimation of PPI extraction performance on new text is possible, given the current heterogeneity in evaluation data. Nevertheless, our study
A One-Sample Test for Normality with Kernel Methods
Kellner , Jérémie; Celisse , Alain
2015-01-01
We propose a new one-sample test for normality in a Reproducing Kernel Hilbert Space (RKHS). Namely, we test the null-hypothesis of belonging to a given family of Gaussian distributions. Hence our procedure may be applied either to test data for normality or to test parameters (mean and covariance) if data are assumed Gaussian. Our test is based on the same principle as the MMD (Maximum Mean Discrepancy) which is usually used for two-sample tests such as homogeneity or independence testing. O...
Regularization method for solving the inverse scattering problem
International Nuclear Information System (INIS)
Denisov, A.M.; Krylov, A.S.
1985-01-01
The inverse scattering problem for the Schroedinger radial equation consisting in determining the potential according to the scattering phase is considered. The problem of potential restoration according to the phase specified with fixed error in a finite range is solved by the regularization method based on minimization of the Tikhonov's smoothing functional. The regularization method is used for solving the problem of neutron-proton potential restoration according to the scattering phases. The determined potentials are given in the table
Schaid, Daniel J
2010-01-01
Measures of genomic similarity are the basis of many statistical analytic methods. We review the mathematical and statistical basis of similarity methods, particularly based on kernel methods. A kernel function converts information for a pair of subjects to a quantitative value representing either similarity (larger values meaning more similar) or distance (smaller values meaning more similar), with the requirement that it must create a positive semidefinite matrix when applied to all pairs of subjects. This review emphasizes the wide range of statistical methods and software that can be used when similarity is based on kernel methods, such as nonparametric regression, linear mixed models and generalized linear mixed models, hierarchical models, score statistics, and support vector machines. The mathematical rigor for these methods is summarized, as is the mathematical framework for making kernels. This review provides a framework to move from intuitive and heuristic approaches to define genomic similarities to more rigorous methods that can take advantage of powerful statistical modeling and existing software. A companion paper reviews novel approaches to creating kernels that might be useful for genomic analyses, providing insights with examples [1]. Copyright © 2010 S. Karger AG, Basel.
Lavrentiev regularization method for nonlinear ill-posed problems
International Nuclear Information System (INIS)
Kinh, Nguyen Van
2002-10-01
In this paper we shall be concerned with Lavientiev regularization method to reconstruct solutions x 0 of non ill-posed problems F(x)=y o , where instead of y 0 noisy data y δ is an element of X with absolut(y δ -y 0 ) ≤ δ are given and F:X→X is an accretive nonlinear operator from a real reflexive Banach space X into itself. In this regularization method solutions x α δ are obtained by solving the singularly perturbed nonlinear operator equation F(x)+α(x-x*)=y δ with some initial guess x*. Assuming certain conditions concerning the operator F and the smoothness of the element x*-x 0 we derive stability estimates which show that the accuracy of the regularized solutions is order optimal provided that the regularization parameter α has been chosen properly. (author)
International Nuclear Information System (INIS)
Takahashi, Akito; Yamamoto, Junji; Ebisuya, Mituo; Sumita, Kenji
1979-01-01
A new method for calculating the anisotropic neutron transport is proposed for the angular spectral analysis of D-T fusion reactor neutronics. The method is based on the transport equation with new type of anisotropic scattering kernels formulated by a single function I sub(i) (μ', μ) instead of polynomial expansion, for instance, Legendre polynomials. In the calculation of angular flux spectra by using scattering kernels with the Legendre polynomial expansion, we often observe the oscillation with negative flux. But in principle this oscillation disappears by this new method. In this work, we discussed anisotropic scattering kernels of the elastic scattering and the inelastic scatterings which excite discrete energy levels. The other scatterings were included in isotropic scattering kernels. An approximation method, with use of the first collision source written by the I sub(i) (μ', μ) function, was introduced to attenuate the ''oscillations'' when we are obliged to use the scattering kernels with the Legendre polynomial expansion. Calculated results with this approximation showed remarkable improvement for the analysis of the angular flux spectra in a slab system of lithium metal with the D-T neutron source. (author)
The regularized monotonicity method: detecting irregular indefinite inclusions
DEFF Research Database (Denmark)
Garde, Henrik; Staboulis, Stratos
2018-01-01
inclusions, where the conductivity distribution has both more and less conductive parts relative to the background conductivity; one such method is the monotonicity method of Harrach, Seo, and Ullrich. We formulate the method for irregular indefinite inclusions, meaning that we make no regularity assumptions...
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.
Directory of Open Access Journals (Sweden)
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.
Reproducing kernel method with Taylor expansion for linear Volterra integro-differential equations
Directory of Open Access Journals (Sweden)
Azizallah Alvandi
2017-06-01
Full Text Available This research aims of the present a new and single algorithm for linear integro-differential equations (LIDE. To apply the reproducing Hilbert kernel method, there is made an equivalent transformation by using Taylor series for solving LIDEs. Shown in series form is the analytical solution in the reproducing kernel space and the approximate solution $ u_{N} $ is constructed by truncating the series to $ N $ terms. It is easy to prove the convergence of $ u_{N} $ to the analytical solution. The numerical solutions from the proposed method indicate that this approach can be implemented easily which shows attractive features.
Development of nondestructive screening methods for single kernel characterization of wheat
DEFF Research Database (Denmark)
Nielsen, J.P.; Pedersen, D.K.; Munck, L.
2003-01-01
predictability. However, by applying an averaging approach, in which single seed replicate measurements are mathematically simulated, a very good NIT prediction model was achieved. This suggests that the single seed NIT spectra contain hardness information, but that a single seed hardness method with higher......The development of nondestructive screening methods for single seed protein, vitreousness, density, and hardness index has been studied for single kernels of European wheat. A single kernel procedure was applied involving, image analysis, near-infrared transmittance (NIT) spectroscopy, laboratory...
Kernel methods and flexible inference for complex stochastic dynamics
Capobianco, Enrico
2008-07-01
Approximation theory suggests that series expansions and projections represent standard tools for random process applications from both numerical and statistical standpoints. Such instruments emphasize the role of both sparsity and smoothness for compression purposes, the decorrelation power achieved in the expansion coefficients space compared to the signal space, and the reproducing kernel property when some special conditions are met. We consider these three aspects central to the discussion in this paper, and attempt to analyze the characteristics of some known approximation instruments employed in a complex application domain such as financial market time series. Volatility models are often built ad hoc, parametrically and through very sophisticated methodologies. But they can hardly deal with stochastic processes with regard to non-Gaussianity, covariance non-stationarity or complex dependence without paying a big price in terms of either model mis-specification or computational efficiency. It is thus a good idea to look at other more flexible inference tools; hence the strategy of combining greedy approximation and space dimensionality reduction techniques, which are less dependent on distributional assumptions and more targeted to achieve computationally efficient performances. Advantages and limitations of their use will be evaluated by looking at algorithmic and model building strategies, and by reporting statistical diagnostics.
A Fourier-series-based kernel-independent fast multipole method
International Nuclear Information System (INIS)
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.
Summation of Divergent Series and Zeldovich's Regularization Method
International Nuclear Information System (INIS)
Mur, V.D.; Pozdnyakov, S.G.; Popruzhenko, S.V.; Popov, V.S.
2005-01-01
A method for summing divergent series, including perturbation-theory series, is considered. This method is an analog of Zeldovich's regularization method in the theory of quasistationary states. It is shown that the method in question is more powerful than the well-known Abel and Borel methods, but that it is compatible with them (that is, it leads to the same value for the sum of a series). The constraints on the parameter domain that arise upon the removal of the regularization of divergent integrals by this method are discussed. The dynamical Stark shifts and widths of loosely bound s states in the field of a circularly polarized electromagnetic wave are calculated at various values of the Keldysh adiabaticity parameter and the multiquantum parameter
REGULARIZED D-BAR METHOD FOR THE INVERSE CONDUCTIVITY PROBLEM
DEFF Research Database (Denmark)
Knudsen, Kim; Lassas, Matti; Mueller, Jennifer
2009-01-01
A strategy for regularizing the inversion procedure for the two-dimensional D-bar reconstruction algorithm based on the global uniqueness proof of Nachman [Ann. Math. 143 (1996)] for the ill-posed inverse conductivity problem is presented. The strategy utilizes truncation of the boundary integral...... the convergence of the reconstructed conductivity to the true conductivity as the noise level tends to zero. The results provide a link between two traditions of inverse problems research: theory of regularization and inversion methods based on complex geometrical optics. Also, the procedure is a novel...
The slab albedo problem for the triplet scattering kernel with modified F{sub N} method
Energy Technology Data Exchange (ETDEWEB)
Tuereci, Demet [Ministry of Education, 75th year Anatolia High School, Ankara (Turkey)
2016-12-15
One speed, time independent neutron transport equation for a slab geometry with the quadratic anisotropic scattering kernel is considered. The albedo and transmission factor are calculated by the modified F{sub N} method. The obtained numerical results are listed for different scattering coefficients.
Biorefinery methods for separation of protein and oil fractions from rubber seed kernel
Widyarani, R.; Ratnaningsih, E.; Sanders, J.P.M.; Bruins, M.E.
2014-01-01
Biorefinery of rubber seeds can generate additional income for farmers, who already grow rubber trees for latex production. The aim of this study was to find the best method for protein and oil production from rubber seed kernel, with focus on protein recovery. Different pre-treatments and oil
Energy Technology Data Exchange (ETDEWEB)
Tuereci, R. Goekhan [Kirikkale Univ. (Turkey). Kirikkale Vocational School; Tuereci, D. [Ministry of Education, Ankara (Turkey). 75th year Anatolia High School
2017-11-15
One speed, time independent and homogeneous medium neutron transport equation is solved with the anisotropic scattering which includes both the linearly and the quadratically anisotropic scattering kernel. Having written Case's eigenfunctions and the orthogonality relations among of these eigenfunctions, slab albedo problem is investigated as numerically by using Modified F{sub N} method. Selected numerical results are presented in tables.
Method of transferring regular shaped vessel into cell
International Nuclear Information System (INIS)
Murai, Tsunehiko.
1997-01-01
The present invention concerns a method of transferring regular shaped vessels from a non-contaminated area to a contaminated cell. A passage hole for allowing the regular shaped vessels to pass in the longitudinal direction is formed to a partitioning wall at the bottom of the contaminated cell. A plurality of regular shaped vessel are stacked in multiple stages in a vertical direction from the non-contaminated area present below the passage hole, allowed to pass while being urged and transferred successively into the contaminated cell. As a result, since they are transferred while substantially closing the passage hole by the regular shaped vessels, radiation rays or contaminated materials are prevented from discharging from the contaminated cell to the non-contaminated area. Since there is no requirement to open/close an isolation door frequently, the workability upon transfer can be improved remarkably. In addition, the sealing member for sealing the gap between the regular shaped vessel passing through the passage hole and the partitioning wall of the bottom is disposed to the passage hole, the contaminated materials in the contaminated cells can be prevented from discharging from the gap to the non-contaminated area. (N.H.)
Improved dynamical scaling analysis using the kernel method for nonequilibrium relaxation.
Echinaka, Yuki; Ozeki, Yukiyasu
2016-10-01
The dynamical scaling analysis for the Kosterlitz-Thouless transition in the nonequilibrium relaxation method is improved by the use of Bayesian statistics and the kernel method. This allows data to be fitted to a scaling function without using any parametric model function, which makes the results more reliable and reproducible and enables automatic and faster parameter estimation. Applying this method, the bootstrap method is introduced and a numerical discrimination for the transition type is proposed.
Summation of divergent series and Zel'dovich's regularization method
International Nuclear Information System (INIS)
Mur, V.D.; Pozdnyakov, S.G.; Popruzhenko, S.V.; Popov, V.S.
2005-01-01
The method of summation of divergent series, including series of a perturbation theory, which is an analog of the Zel'dovich regularization procedure in the theory of quasistationary states is considered. It is shown that this method is more powerful than the well-known Abel and Borel methods, but compatible with them (i. e., gives the same value for the sum of the series). The restrictions to the range of parameters which appear after removal of the regularization of integrals by this method are discussed. The dynamical Stark shifts and widths of weakly bound s states in a field of circularly polarized electromagnetic wave are calculated at different values of the Keldysh adiabaticity parameter and multiquantum parameter [ru
The Validity of Dimensional Regularization Method on Fractal Spacetime
Directory of Open Access Journals (Sweden)
Yong Tao
2013-01-01
Full Text Available Svozil developed a regularization method for quantum field theory on fractal spacetime (1987. Such a method can be applied to the low-order perturbative renormalization of quantum electrodynamics but will depend on a conjectural integral formula on non-integer-dimensional topological spaces. The main purpose of this paper is to construct a fractal measure so as to guarantee the validity of the conjectural integral formula.
Regularization of the double period method for experimental data processing
Belov, A. A.; Kalitkin, N. N.
2017-11-01
In physical and technical applications, an important task is to process experimental curves measured with large errors. Such problems are solved by applying regularization methods, in which success depends on the mathematician's intuition. We propose an approximation based on the double period method developed for smooth nonperiodic functions. Tikhonov's stabilizer with a squared second derivative is used for regularization. As a result, the spurious oscillations are suppressed and the shape of an experimental curve is accurately represented. This approach offers a universal strategy for solving a broad class of problems. The method is illustrated by approximating cross sections of nuclear reactions important for controlled thermonuclear fusion. Tables recommended as reference data are obtained. These results are used to calculate the reaction rates, which are approximated in a way convenient for gasdynamic codes. These approximations are superior to previously known formulas in the covered temperature range and accuracy.
Use of regularized algebraic methods in tomographic reconstruction
International Nuclear Information System (INIS)
Koulibaly, P.M.; Darcourt, J.; Blanc-Ferraud, L.; Migneco, O.; Barlaud, M.
1997-01-01
The algebraic methods are used in emission tomography to facilitate the compensation of attenuation and of Compton scattering. We have tested on a phantom the use of a regularization (a priori introduction of information), as well as the taking into account of spatial resolution variation with the depth (SRVD). Hence, we have compared the performances of the two methods by back-projection filtering (BPF) and of the two algebraic methods (AM) in terms of FWHM (by means of a point source), of the reduction of background noise (σ/m) on the homogeneous part of Jaszczak's phantom and of reconstruction speed (time unit = BPF). The BPF methods make use of a grade filter (maximal resolution, no noise treatment), single or associated with a Hann's low-pass (f c = 0.4), as well as of an attenuation correction. The AM which embody attenuation and scattering corrections are, on one side, the OS EM (Ordered Subsets, partitioning and rearranging of the projection matrix; Expectation Maximization) without regularization or SRVD correction, and, on the other side, the OS MAP EM (Maximum a posteriori), regularized and embodying the SRVD correction. A table is given containing for each used method (grade, Hann, OS EM and OS MAP EM) the values of FWHM, σ/m and time, respectively. One can observe that the OS MAP EM algebraic method allows ameliorating both the resolution, by taking into account the SRVD in the reconstruction process and noise treatment by regularization. In addition, due to the OS technique the reconstruction times are acceptable
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.
Regularization by fractional filter methods and data smoothing
International Nuclear Information System (INIS)
Klann, E; Ramlau, R
2008-01-01
This paper is concerned with the regularization of linear ill-posed problems by a combination of data smoothing and fractional filter methods. For the data smoothing, a wavelet shrinkage denoising is applied to the noisy data with known error level δ. For the reconstruction, an approximation to the solution of the operator equation is computed from the data estimate by fractional filter methods. These fractional methods are based on the classical Tikhonov and Landweber method, but avoid, at least partially, the well-known drawback of oversmoothing. Convergence rates as well as numerical examples are presented
International Nuclear Information System (INIS)
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)
A Novel Mittag-Leffler Kernel Based Hybrid Fault Diagnosis Method for Wheeled Robot Driving System
Directory of Open Access Journals (Sweden)
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.
A two-way regularization method for MEG source reconstruction
Tian, Tian Siva; Huang, Jianhua Z.; Shen, Haipeng; Li, Zhimin
2012-01-01
The MEG inverse problem refers to the reconstruction of the neural activity of the brain from magnetoencephalography (MEG) measurements. We propose a two-way regularization (TWR) method to solve the MEG inverse problem under the assumptions that only a small number of locations in space are responsible for the measured signals (focality), and each source time course is smooth in time (smoothness). The focality and smoothness of the reconstructed signals are ensured respectively by imposing a sparsity-inducing penalty and a roughness penalty in the data fitting criterion. A two-stage algorithm is developed for fast computation, where a raw estimate of the source time course is obtained in the first stage and then refined in the second stage by the two-way regularization. The proposed method is shown to be effective on both synthetic and real-world examples. © Institute of Mathematical Statistics, 2012.
A regularization method for extrapolation of solar potential magnetic fields
Gary, G. A.; Musielak, Z. E.
1992-01-01
The mathematical basis of a Tikhonov regularization method for extrapolating the chromospheric-coronal magnetic field using photospheric vector magnetograms is discussed. The basic techniques show that the Cauchy initial value problem can be formulated for potential magnetic fields. The potential field analysis considers a set of linear, elliptic partial differential equations. It is found that, by introducing an appropriate smoothing of the initial data of the Cauchy potential problem, an approximate Fourier integral solution is found, and an upper bound to the error in the solution is derived. This specific regularization technique, which is a function of magnetograph measurement sensitivities, provides a method to extrapolate the potential magnetic field above an active region into the chromosphere and low corona.
A two-way regularization method for MEG source reconstruction
Tian, Tian Siva
2012-09-01
The MEG inverse problem refers to the reconstruction of the neural activity of the brain from magnetoencephalography (MEG) measurements. We propose a two-way regularization (TWR) method to solve the MEG inverse problem under the assumptions that only a small number of locations in space are responsible for the measured signals (focality), and each source time course is smooth in time (smoothness). The focality and smoothness of the reconstructed signals are ensured respectively by imposing a sparsity-inducing penalty and a roughness penalty in the data fitting criterion. A two-stage algorithm is developed for fast computation, where a raw estimate of the source time course is obtained in the first stage and then refined in the second stage by the two-way regularization. The proposed method is shown to be effective on both synthetic and real-world examples. © Institute of Mathematical Statistics, 2012.
Sound Attenuation in Elliptic Mufflers Using a Regular Perturbation Method
Banerjee, Subhabrata; Jacobi, Anthony M.
2012-01-01
The study of sound attenuation in an elliptical chamber involves the solution of the Helmholtz equation in elliptic coordinate systems. The Eigen solutions for such problems involve the Mathieu and the modified Mathieu functions. The computation of such functions poses considerable challenge. An alternative method to solve such problems had been proposed in this paper. The elliptical cross-section of the muffler has been treated as a perturbed circle, enabling the use of a regular perturbatio...
Global regularization method for planar restricted three-body problem
Directory of Open Access Journals (Sweden)
Sharaf M.A.
2015-01-01
Full Text Available In this paper, global regularization method for planar restricted three-body problem is purposed by using the transformation z = x+iy = ν cos n(u+iv, where i = √−1, 0 < ν ≤ 1 and n is a positive integer. The method is developed analytically and computationally. For the analytical developments, analytical solutions in power series of the pseudotime τ are obtained for positions and velocities (u, v, u', v' and (x, y, x˙, y˙ in both regularized and physical planes respectively, the physical time t is also obtained as power series in τ. Moreover, relations between the coefficients of the power series are obtained for two consequent values of n. Also, we developed analytical solutions in power series form for the inverse problem of finding τ in terms of t. As typical examples, three symbolic expressions for the coefficients of the power series were developed in terms of initial values. As to the computational developments, the global regularized equations of motion are developed together with their initial values in forms suitable for digital computations using any differential equations solver. On the other hand, for numerical evolutions of power series, an efficient method depending on the continued fraction theory is provided.
The CRF-method for semiconductors' intravalley collision kernels: I – the 2D case
Directory of Open Access Journals (Sweden)
Claudio Barone
1992-05-01
Full Text Available If the collisions are redefined as a flux a kinetic conservation law can be written in divergence form. This can be handled numerically, in the framework of Finite Particle Approximation, using the CRF-method. In the present paper the relevant quantities needed for computer implementation of the CRF-method are derived in the case of a 2D momentum space for the semiconductors' intravalley collision kernels.
The CRF-method for semiconductors' intravalley collision kernels: II – The 3D case
Directory of Open Access Journals (Sweden)
Claudio Barone
1993-05-01
Full Text Available If the collisions are redefined as a flux a kinetic conservation law can be written in divergence form. This can be handled numerically, in the framework of Finite Particle Approximation, using the CRF-method. In this paper we use the CRF-method for semiconductors' intravalley collision kernels. We extend the results obtained in a previous paper to the case of a 3D momentum space.
A kernel plus method for quantifying wind turbine performance upgrades
Lee, Giwhyun; Ding, Yu; Xie, Le; Genton, Marc G.
2014-01-01
Power curves are commonly estimated using the binning method recommended by the International Electrotechnical Commission, which primarily incorporates wind speed information. When such power curves are used to quantify a turbine's upgrade
A Spectral-Texture Kernel-Based Classification Method for Hyperspectral Images
Directory of Open Access Journals (Sweden)
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.
WASTK: A Weighted Abstract Syntax Tree Kernel Method for Source Code Plagiarism Detection
Directory of Open Access Journals (Sweden)
Deqiang Fu
2017-01-01
Full Text Available In this paper, we introduce a source code plagiarism detection method, named WASTK (Weighted Abstract Syntax Tree Kernel, for computer science education. Different from other plagiarism detection methods, WASTK takes some aspects other than the similarity between programs into account. WASTK firstly transfers the source code of a program to an abstract syntax tree and then gets the similarity by calculating the tree kernel of two abstract syntax trees. To avoid misjudgment caused by trivial code snippets or frameworks given by instructors, an idea similar to TF-IDF (Term Frequency-Inverse Document Frequency in the field of information retrieval is applied. Each node in an abstract syntax tree is assigned a weight by TF-IDF. WASTK is evaluated on different datasets and, as a result, performs much better than other popular methods like Sim and JPlag.
Directory of Open Access Journals (Sweden)
Omar Abu Arqub
2012-01-01
Full Text Available This paper investigates the numerical solution of nonlinear Fredholm-Volterra integro-differential equations using reproducing kernel Hilbert space method. The solution ( is represented in the form of series in the reproducing kernel space. In the mean time, the n-term approximate solution ( is obtained and it is proved to converge to the exact solution (. Furthermore, the proposed method has an advantage that it is possible to pick any point in the interval of integration and as well the approximate solution and its derivative will be applicable. Numerical examples are included to demonstrate the accuracy and applicability of the presented technique. The results reveal that the method is very effective and simple.
Scalable Kernel Methods and Algorithms for General Sequence Analysis
Kuksa, Pavel
2011-01-01
Analysis of large-scale sequential data has become an important task in machine learning and pattern recognition, inspired in part by numerous scientific and technological applications such as the document and text classification or the analysis of biological sequences. However, current computational methods for sequence comparison still lack…
GLOBAL OPTIMIZATION METHODS FOR GRAVITATIONAL LENS SYSTEMS WITH REGULARIZED SOURCES
International Nuclear Information System (INIS)
Rogers, Adam; Fiege, Jason D.
2012-01-01
Several approaches exist to model gravitational lens systems. In this study, we apply global optimization methods to find the optimal set of lens parameters using a genetic algorithm. We treat the full optimization procedure as a two-step process: an analytical description of the source plane intensity distribution is used to find an initial approximation to the optimal lens parameters; the second stage of the optimization uses a pixelated source plane with the semilinear method to determine an optimal source. Regularization is handled by means of an iterative method and the generalized cross validation (GCV) and unbiased predictive risk estimator (UPRE) functions that are commonly used in standard image deconvolution problems. This approach simultaneously estimates the optimal regularization parameter and the number of degrees of freedom in the source. Using the GCV and UPRE functions, we are able to justify an estimation of the number of source degrees of freedom found in previous work. We test our approach by applying our code to a subset of the lens systems included in the SLACS survey.
Kernel machine methods for integrative analysis of genome-wide methylation and genotyping studies.
Zhao, Ni; Zhan, Xiang; Huang, Yen-Tsung; Almli, Lynn M; Smith, Alicia; Epstein, Michael P; Conneely, Karen; Wu, Michael C
2018-03-01
Many large GWAS consortia are expanding to simultaneously examine the joint role of DNA methylation in addition to genotype in the same subjects. However, integrating information from both data types is challenging. In this paper, we propose a composite kernel machine regression model to test the joint epigenetic and genetic effect. Our approach works at the gene level, which allows for a common unit of analysis across different data types. The model compares the pairwise similarities in the phenotype to the pairwise similarities in the genotype and methylation values; and high correspondence is suggestive of association. A composite kernel is constructed to measure the similarities in the genotype and methylation values between pairs of samples. We demonstrate through simulations and real data applications that the proposed approach can correctly control type I error, and is more robust and powerful than using only the genotype or methylation data in detecting trait-associated genes. We applied our method to investigate the genetic and epigenetic regulation of gene expression in response to stressful life events using data that are collected from the Grady Trauma Project. Within the kernel machine testing framework, our methods allow for heterogeneity in effect sizes, nonlinear, and interactive effects, as well as rapid P-value computation. © 2017 WILEY PERIODICALS, INC.
A new method by steering kernel-based Richardson–Lucy algorithm for neutron imaging restoration
International Nuclear Information System (INIS)
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
Generation of gamma-ray streaming kernels through cylindrical ducts via Monte Carlo method
International Nuclear Information System (INIS)
Kim, Dong Su
1992-02-01
Since radiation streaming through penetrations is often the critical consideration in protection against exposure of personnel in a nuclear facility, it has been of great concern in radiation shielding design and analysis. Several methods have been developed and applied to the analysis of the radiation streaming in the past such as ray analysis method, single scattering method, albedo method, and Monte Carlo method. But they may be used for order-of-magnitude calculations and where sufficient margin is available, except for the Monte Carlo method which is accurate but requires a lot of computing time. This study developed a Monte Carlo method and constructed a data library of solutions using the Monte Carlo method for radiation streaming through a straight cylindrical duct in concrete walls of a broad, mono-directional, monoenergetic gamma-ray beam of unit intensity. The solution named as plane streaming kernel is the average dose rate at duct outlet and was evaluated for 20 source energies from 0 to 10 MeV, 36 source incident angles from 0 to 70 degrees, 5 duct radii from 10 to 30 cm, and 16 wall thicknesses from 0 to 100 cm. It was demonstrated that average dose rate due to an isotropic point source at arbitrary positions can be well approximated using the plane streaming kernel with acceptable error. Thus, the library of the plane streaming kernels can be used for the accurate and efficient analysis of radiation streaming through a straight cylindrical duct in concrete walls due to arbitrary distributions of gamma-ray sources
AbdulJabbar, Mustafa Abdulmajeed
2017-07-31
Manycore optimizations are essential for achieving performance worthy of anticipated exascale systems. Utilization of manycore chips is inevitable to attain the desired floating point performance of these energy-austere systems. In this work, we revisit ExaFMM, the open source Fast Multiple Method (FMM) library, in light of highly tuned shared-memory parallelization and detailed performance analysis on the new highly parallel Intel manycore architecture, Knights Landing (KNL). We assess scalability and performance gain using task-based parallelism of the FMM tree traversal. We also provide an in-depth analysis of the most computationally intensive part of the traversal kernel (i.e., the particle-to-particle (P2P) kernel), by comparing its performance across KNL and Broadwell architectures. We quantify different configurations that exploit the on-chip 512-bit vector units within different task-based threading paradigms. MPI communication-reducing and NUMA-aware approaches for the FMM’s global tree data exchange are examined with different cluster modes of KNL. By applying several algorithm- and architecture-aware optimizations for FMM, we show that the N-Body kernel on 256 threads of KNL achieves on average 2.8× speedup compared to the non-vectorized version, whereas on 56 threads of Broadwell, it achieves on average 2.9× speedup. In addition, the tree traversal kernel on KNL scales monotonically up to 256 threads with task-based programming models. The MPI-based communication-reducing algorithms show expected improvements of the data locality across the KNL on-chip network.
AbdulJabbar, Mustafa Abdulmajeed; Al Farhan, Mohammed; Yokota, Rio; Keyes, David E.
2017-01-01
Manycore optimizations are essential for achieving performance worthy of anticipated exascale systems. Utilization of manycore chips is inevitable to attain the desired floating point performance of these energy-austere systems. In this work, we revisit ExaFMM, the open source Fast Multiple Method (FMM) library, in light of highly tuned shared-memory parallelization and detailed performance analysis on the new highly parallel Intel manycore architecture, Knights Landing (KNL). We assess scalability and performance gain using task-based parallelism of the FMM tree traversal. We also provide an in-depth analysis of the most computationally intensive part of the traversal kernel (i.e., the particle-to-particle (P2P) kernel), by comparing its performance across KNL and Broadwell architectures. We quantify different configurations that exploit the on-chip 512-bit vector units within different task-based threading paradigms. MPI communication-reducing and NUMA-aware approaches for the FMM’s global tree data exchange are examined with different cluster modes of KNL. By applying several algorithm- and architecture-aware optimizations for FMM, we show that the N-Body kernel on 256 threads of KNL achieves on average 2.8× speedup compared to the non-vectorized version, whereas on 56 threads of Broadwell, it achieves on average 2.9× speedup. In addition, the tree traversal kernel on KNL scales monotonically up to 256 threads with task-based programming models. The MPI-based communication-reducing algorithms show expected improvements of the data locality across the KNL on-chip network.
Recursive regularization step for high-order lattice Boltzmann methods
Coreixas, Christophe; Wissocq, Gauthier; Puigt, Guillaume; Boussuge, Jean-François; Sagaut, Pierre
2017-09-01
A lattice Boltzmann method (LBM) with enhanced stability and accuracy is presented for various Hermite tensor-based lattice structures. The collision operator relies on a regularization step, which is here improved through a recursive computation of nonequilibrium Hermite polynomial coefficients. In addition to the reduced computational cost of this procedure with respect to the standard one, the recursive step allows to considerably enhance the stability and accuracy of the numerical scheme by properly filtering out second- (and higher-) order nonhydrodynamic contributions in under-resolved conditions. This is first shown in the isothermal case where the simulation of the doubly periodic shear layer is performed with a Reynolds number ranging from 104 to 106, and where a thorough analysis of the case at Re=3 ×104 is conducted. In the latter, results obtained using both regularization steps are compared against the Bhatnagar-Gross-Krook LBM for standard (D2Q9) and high-order (D2V17 and D2V37) lattice structures, confirming the tremendous increase of stability range of the proposed approach. Further comparisons on thermal and fully compressible flows, using the general extension of this procedure, are then conducted through the numerical simulation of Sod shock tubes with the D2V37 lattice. They confirm the stability increase induced by the recursive approach as compared with the standard one.
Directory of Open Access Journals (Sweden)
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.
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.
Kernel polynomial method for a nonorthogonal electronic-structure calculation of amorphous diamond
International Nuclear Information System (INIS)
Roeder, H.; Silver, R.N.; Drabold, D.A.; Dong, J.J.
1997-01-01
The Kernel polynomial method (KPM) has been successfully applied to tight-binding electronic-structure calculations as an O(N) method. Here we extend this method to nonorthogonal basis sets with a sparse overlap matrix S and a sparse Hamiltonian H. Since the KPM method utilizes matrix vector multiplications it is necessary to apply S -1 H onto a vector. The multiplication of S -1 is performed using a preconditioned conjugate-gradient method and does not involve the explicit inversion of S. Hence the method scales the same way as the original KPM method, i.e., O(N), although there is an overhead due to the additional conjugate-gradient part. We apply this method to a large scale electronic-structure calculation of amorphous diamond. copyright 1997 The American Physical Society
Novel applications of the temporal kernel method: Historical and future radiative forcing
Portmann, R. W.; Larson, E.; Solomon, S.; Murphy, D. M.
2017-12-01
We present a new estimate of the historical radiative forcing derived from the observed global mean surface temperature and a model derived kernel function. Current estimates of historical radiative forcing are usually derived from climate models. Despite large variability in these models, the multi-model mean tends to do a reasonable job of representing the Earth system and climate. One method of diagnosing the transient radiative forcing in these models requires model output of top of the atmosphere radiative imbalance and global mean temperature anomaly. It is difficult to apply this method to historical observations due to the lack of TOA radiative measurements before CERES. We apply the temporal kernel method (TKM) of calculating radiative forcing to the historical global mean temperature anomaly. This novel approach is compared against the current regression based methods using model outputs and shown to produce consistent forcing estimates giving confidence in the forcing derived from the historical temperature record. The derived TKM radiative forcing provides an estimate of the forcing time series that the average climate model needs to produce the observed temperature record. This forcing time series is found to be in good overall agreement with previous estimates but includes significant differences that will be discussed. The historical anthropogenic aerosol forcing is estimated as a residual from the TKM and found to be consistent with earlier moderate forcing estimates. In addition, this method is applied to future temperature projections to estimate the radiative forcing required to achieve those temperature goals, such as those set in the Paris agreement.
Directory of Open Access Journals (Sweden)
ELİF BULUT
2013-06-01
Full Text Available Partial Least Squares Regression (PLSR is a multivariate statistical method that consists of partial least squares and multiple linear regression analysis. Explanatory variables, X, having multicollinearity are reduced to components which explain the great amount of covariance between explanatory and response variable. These components are few in number and they don’t have multicollinearity problem. Then multiple linear regression analysis is applied to those components to model the response variable Y. There are various PLSR algorithms. In this study NIPALS and PLS-Kernel algorithms will be studied and illustrated on a real data set.
Kernel method for air quality modelling. II. Comparison with analytic solutions
Energy Technology Data Exchange (ETDEWEB)
Lorimer, G S; Ross, D G
1986-01-01
The performance of Lorimer's (1986) kernel method for solving the advection-diffusion equation is tested for instantaneous and continuous emissions into a variety of model atmospheres. Analytical solutions are available for comparison in each case. The results indicate that a modest minicomputer is quite adequate for obtaining satisfactory precision even for the most trying test performed here, which involves a diffusivity tensor and wind speed which are nonlinear functions of the height above ground. Simulations of the same cases by the particle-in-cell technique are found to provide substantially lower accuracy even when use is made of greater computer resources.
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.
Directory of Open Access Journals (Sweden)
Ji Li
2016-10-01
Full Text Available A piezo-resistive pressure sensor is made of silicon, the nature of which is considerably influenced by ambient temperature. The effect of temperature should be eliminated during the working period in expectation of linear output. To deal with this issue, an approach consists of a hybrid kernel Least Squares Support Vector Machine (LSSVM optimized by a chaotic ions motion algorithm presented. To achieve the learning and generalization for excellent performance, a hybrid kernel function, constructed by a local kernel as Radial Basis Function (RBF kernel, and a global kernel as polynomial kernel is incorporated into the Least Squares Support Vector Machine. The chaotic ions motion algorithm is introduced to find the best hyper-parameters of the Least Squares Support Vector Machine. The temperature data from a calibration experiment is conducted to validate the proposed method. With attention on algorithm robustness and engineering applications, the compensation result shows the proposed scheme outperforms other compared methods on several performance measures as maximum absolute relative error, minimum absolute relative error mean and variance of the averaged value on fifty runs. Furthermore, the proposed temperature compensation approach lays a foundation for more extensive research.
Li, Ji; Hu, Guoqing; Zhou, Yonghong; Zou, Chong; Peng, Wei; Alam Sm, Jahangir
2016-10-14
A piezo-resistive pressure sensor is made of silicon, the nature of which is considerably influenced by ambient temperature. The effect of temperature should be eliminated during the working period in expectation of linear output. To deal with this issue, an approach consists of a hybrid kernel Least Squares Support Vector Machine (LSSVM) optimized by a chaotic ions motion algorithm presented. To achieve the learning and generalization for excellent performance, a hybrid kernel function, constructed by a local kernel as Radial Basis Function (RBF) kernel, and a global kernel as polynomial kernel is incorporated into the Least Squares Support Vector Machine. The chaotic ions motion algorithm is introduced to find the best hyper-parameters of the Least Squares Support Vector Machine. The temperature data from a calibration experiment is conducted to validate the proposed method. With attention on algorithm robustness and engineering applications, the compensation result shows the proposed scheme outperforms other compared methods on several performance measures as maximum absolute relative error, minimum absolute relative error mean and variance of the averaged value on fifty runs. Furthermore, the proposed temperature compensation approach lays a foundation for more extensive research.
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.
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
A kernel-based multivariate feature selection method for microarray data classification.
Directory of Open Access Journals (Sweden)
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.
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.
Nonparametric Inference of Doubly Stochastic Poisson Process Data via the Kernel Method.
Zhang, Tingting; Kou, S C
2010-01-01
Doubly stochastic Poisson processes, also known as the Cox processes, frequently occur in various scientific fields. In this article, motivated primarily by analyzing Cox process data in biophysics, we propose a nonparametric kernel-based inference method. We conduct a detailed study, including an asymptotic analysis, of the proposed method, and provide guidelines for its practical use, introducing a fast and stable regression method for bandwidth selection. We apply our method to real photon arrival data from recent single-molecule biophysical experiments, investigating proteins' conformational dynamics. Our result shows that conformational fluctuation is widely present in protein systems, and that the fluctuation covers a broad range of time scales, highlighting the dynamic and complex nature of proteins' structure.
MR-guided dynamic PET reconstruction with the kernel method and spectral temporal basis functions
Novosad, Philip; Reader, Andrew J.
2016-06-01
Recent advances in dynamic positron emission tomography (PET) reconstruction have demonstrated that it is possible to achieve markedly improved end-point kinetic parameter maps by incorporating a temporal model of the radiotracer directly into the reconstruction algorithm. In this work we have developed a highly constrained, fully dynamic PET reconstruction algorithm incorporating both spectral analysis temporal basis functions and spatial basis functions derived from the kernel method applied to a co-registered T1-weighted magnetic resonance (MR) image. The dynamic PET image is modelled as a linear combination of spatial and temporal basis functions, and a maximum likelihood estimate for the coefficients can be found using the expectation-maximization (EM) algorithm. Following reconstruction, kinetic fitting using any temporal model of interest can be applied. Based on a BrainWeb T1-weighted MR phantom, we performed a realistic dynamic [18F]FDG simulation study with two noise levels, and investigated the quantitative performance of the proposed reconstruction algorithm, comparing it with reconstructions incorporating either spectral analysis temporal basis functions alone or kernel spatial basis functions alone, as well as with conventional frame-independent reconstruction. Compared to the other reconstruction algorithms, the proposed algorithm achieved superior performance, offering a decrease in spatially averaged pixel-level root-mean-square-error on post-reconstruction kinetic parametric maps in the grey/white matter, as well as in the tumours when they were present on the co-registered MR image. When the tumours were not visible in the MR image, reconstruction with the proposed algorithm performed similarly to reconstruction with spectral temporal basis functions and was superior to both conventional frame-independent reconstruction and frame-independent reconstruction with kernel spatial basis functions. Furthermore, we demonstrate that a joint spectral/kernel
International Nuclear Information System (INIS)
Jin Qinian
2008-01-01
In this paper we consider the iteratively regularized Gauss–Newton method for solving nonlinear ill-posed inverse problems. Under merely the Lipschitz condition, we prove that this method together with an a posteriori stopping rule defines an order optimal regularization method if the solution is regular in some suitable sense
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 ...
Improved Expectation Maximization Algorithm for Gaussian Mixed Model Using the Kernel Method
Directory of Open Access Journals (Sweden)
Mohd Izhan Mohd Yusoff
2013-01-01
Full Text Available Fraud activities have contributed to heavy losses suffered by telecommunication companies. In this paper, we attempt to use Gaussian mixed model, which is a probabilistic model normally used in speech recognition to identify fraud calls in the telecommunication industry. We look at several issues encountered when calculating the maximum likelihood estimates of the Gaussian mixed model using an Expectation Maximization algorithm. Firstly, we look at a mechanism for the determination of the initial number of Gaussian components and the choice of the initial values of the algorithm using the kernel method. We show via simulation that the technique improves the performance of the algorithm. Secondly, we developed a procedure for determining the order of the Gaussian mixed model using the log-likelihood function and the Akaike information criteria. Finally, for illustration, we apply the improved algorithm to real telecommunication data. The modified method will pave the way to introduce a comprehensive method for detecting fraud calls in future work.
Energy Technology Data Exchange (ETDEWEB)
Jeong, Kyung Chai; Eom, Sung Ho; Kim, Yeon Ku; Yeo, Seoung Hwan; Kim, Young Min; Cho, Moon Sung [Korea Atomic Energy Research Institute, Daejeon (Korea, Republic of)
2016-10-15
VHTR (Very High Temperature Gas Reactor) fuel technology is being actively developed in the US, China, Japan, and Korea for a Next Generation Nuclear Plant (NGNP). The concept of fuel of a VHTR is based on a sphere kernel of UO{sub 2} or UCO, with multiple coating layers to create a gas-tight particle. The fuel particle of a VHTR in the US is based on microspheres containing a UCO, mixture compound of UO{sub 2} and UC{sub 2} , coated particles with multi carbon layers and a SiC layer. This was first prepared through an internal gelation method at ORNL in the late 1970s. This study presents; (1) C-ADU gel particles were prepared using a modified sol-gel process. The particles fabricated with a KAERI-established gelation and AWD processes showed good sphericity and no cracks were found on the surfaces. (2) High temperature rotating furnace was designed and fabricated in our laboratory, and the maximum operation temperature was about 2000℃. The furnace was equipped with Mo crucible and graphite heating system, and now it is being operated. (3) Well-prepared C-ADU gel particles were converted into UCO compounds using high temperature rotating furnace, and the physical properties of the UCO kernels will be analyzed.
Analytical Plug-In Method for Kernel Density Estimator Applied to Genetic Neutrality Study
Troudi, Molka; Alimi, Adel M.; Saoudi, Samir
2008-12-01
The plug-in method enables optimization of the bandwidth of the kernel density estimator in order to estimate probability density functions (pdfs). Here, a faster procedure than that of the common plug-in method is proposed. The mean integrated square error (MISE) depends directly upon [InlineEquation not available: see fulltext.] which is linked to the second-order derivative of the pdf. As we intend to introduce an analytical approximation of [InlineEquation not available: see fulltext.], the pdf is estimated only once, at the end of iterations. These two kinds of algorithm are tested on different random variables having distributions known for their difficult estimation. Finally, they are applied to genetic data in order to provide a better characterisation in the mean of neutrality of Tunisian Berber populations.
Analytical Plug-In Method for Kernel Density Estimator Applied to Genetic Neutrality Study
Directory of Open Access Journals (Sweden)
Samir Saoudi
2008-07-01
Full Text Available The plug-in method enables optimization of the bandwidth of the kernel density estimator in order to estimate probability density functions (pdfs. Here, a faster procedure than that of the common plug-in method is proposed. The mean integrated square error (MISE depends directly upon J(f which is linked to the second-order derivative of the pdf. As we intend to introduce an analytical approximation of J(f, the pdf is estimated only once, at the end of iterations. These two kinds of algorithm are tested on different random variables having distributions known for their difficult estimation. Finally, they are applied to genetic data in order to provide a better characterisation in the mean of neutrality of Tunisian Berber populations.
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.
International Nuclear Information System (INIS)
Ducru, Pablo; Josey, Colin; Dibert, Karia; Sobes, Vladimir; Forget, Benoit; Smith, Kord
2017-01-01
This paper establishes a new family of methods to perform temperature interpolation of nuclear interactions cross sections, reaction rates, or cross sections times the energy. One of these quantities at temperature T is approximated as a linear combination of quantities at reference temperatures (T_j). The problem is formalized in a cross section independent fashion by considering the kernels of the different operators that convert cross section related quantities from a temperature T_0 to a higher temperature T — namely the Doppler broadening operation. Doppler broadening interpolation of nuclear cross sections is thus here performed by reconstructing the kernel of the operation at a given temperature T by means of linear combination of kernels at reference temperatures (T_j). The choice of the L_2 metric yields optimal linear interpolation coefficients in the form of the solutions of a linear algebraic system inversion. The optimization of the choice of reference temperatures (T_j) is then undertaken so as to best reconstruct, in the L∞ sense, the kernels over a given temperature range [T_m_i_n,T_m_a_x]. The performance of these kernel reconstruction methods is then assessed in light of previous temperature interpolation methods by testing them upon isotope "2"3"8U. Temperature-optimized free Doppler kernel reconstruction significantly outperforms all previous interpolation-based methods, achieving 0.1% relative error on temperature interpolation of "2"3"8U total cross section over the temperature range [300 K,3000 K] with only 9 reference temperatures.
The Use of Nonparametric Kernel Regression Methods in Econometric Production Analysis
DEFF Research Database (Denmark)
Czekaj, Tomasz Gerard
and nonparametric estimations of production functions in order to evaluate the optimal firm size. The second paper discusses the use of parametric and nonparametric regression methods to estimate panel data regression models. The third paper analyses production risk, price uncertainty, and farmers' risk preferences...... within a nonparametric panel data regression framework. The fourth paper analyses the technical efficiency of dairy farms with environmental output using nonparametric kernel regression in a semiparametric stochastic frontier analysis. The results provided in this PhD thesis show that nonparametric......This PhD thesis addresses one of the fundamental problems in applied econometric analysis, namely the econometric estimation of regression functions. The conventional approach to regression analysis is the parametric approach, which requires the researcher to specify the form of the regression...
Grant, Mary C.; Zhang, Lilly; Damiano, Michele
2009-01-01
This study investigated kernel equating methods by comparing these methods to operational equatings for two tests in the SAT Subject Tests[TM] program. GENASYS (ETS, 2007) was used for all equating methods and scaled score kernel equating results were compared to Tucker, Levine observed score, chained linear, and chained equipercentile equating…
International Nuclear Information System (INIS)
Li Heng; Mohan, Radhe; Zhu, X Ronald
2008-01-01
The clinical applications of kilovoltage x-ray cone-beam computed tomography (CBCT) have been compromised by the limited quality of CBCT images, which typically is due to a substantial scatter component in the projection data. In this paper, we describe an experimental method of deriving the scatter kernel of a CBCT imaging system. The estimated scatter kernel can be used to remove the scatter component from the CBCT projection images, thus improving the quality of the reconstructed image. The scattered radiation was approximated as depth-dependent, pencil-beam kernels, which were derived using an edge-spread function (ESF) method. The ESF geometry was achieved with a half-beam block created by a 3 mm thick lead sheet placed on a stack of slab solid-water phantoms. Measurements for ten water-equivalent thicknesses (WET) ranging from 0 cm to 41 cm were taken with (half-blocked) and without (unblocked) the lead sheet, and corresponding pencil-beam scatter kernels or point-spread functions (PSFs) were then derived without assuming any empirical trial function. The derived scatter kernels were verified with phantom studies. Scatter correction was then incorporated into the reconstruction process to improve image quality. For a 32 cm diameter cylinder phantom, the flatness of the reconstructed image was improved from 22% to 5%. When the method was applied to CBCT images for patients undergoing image-guided therapy of the pelvis and lung, the variation in selected regions of interest (ROIs) was reduced from >300 HU to <100 HU. We conclude that the scatter reduction technique utilizing the scatter kernel effectively suppresses the artifact caused by scatter in CBCT.
Training Lp norm multiple kernel learning in the primal.
Liang, Zhizheng; Xia, Shixiong; Zhou, Yong; Zhang, Lei
2013-10-01
Some multiple kernel learning (MKL) models are usually solved by utilizing the alternating optimization method where one alternately solves SVMs in the dual and updates kernel weights. Since the dual and primal optimization can achieve the same aim, it is valuable in exploring how to perform Lp norm MKL in the primal. In this paper, we propose an Lp norm multiple kernel learning algorithm in the primal where we resort to the alternating optimization method: one cycle for solving SVMs in the primal by using the preconditioned conjugate gradient method and other cycle for learning the kernel weights. It is interesting to note that the kernel weights in our method can obtain analytical solutions. Most importantly, the proposed method is well suited for the manifold regularization framework in the primal since solving LapSVMs in the primal is much more effective than solving LapSVMs in the dual. In addition, we also carry out theoretical analysis for multiple kernel learning in the primal in terms of the empirical Rademacher complexity. It is found that optimizing the empirical Rademacher complexity may obtain a type of kernel weights. The experiments on some datasets are carried out to demonstrate the feasibility and effectiveness of the proposed method. Copyright © 2013 Elsevier Ltd. All rights reserved.
Accuracy of approximations of solutions to Fredholm equations by kernel methods
Czech Academy of Sciences Publication Activity Database
Gnecco, G.; Kůrková, Věra; Sanguineti, M.
2012-01-01
Roč. 218, č. 14 (2012), s. 7481-7497 ISSN 0096-3003 R&D Projects: GA ČR GAP202/11/1368; GA MŠk OC10047 Grant - others:CNR-AV ČR(CZ-IT) Project 2010–2012 “Complexity of Neural -Network and Kernel Computational Models Institutional research plan: CEZ:AV0Z10300504 Keywords : approximate solutions to integral equations * radial and kernel-based networks * Gaussian kernels * model complexity * analysis of algorithms Subject RIV: IN - Informatics, Computer Science Impact factor: 1.349, year: 2012
Directory of Open Access Journals (Sweden)
Wei Gao
2016-01-01
Full Text Available According to the regularization method in the inverse problem of load identification, a new method for determining the optimal regularization parameter is proposed. Firstly, quotient function (QF is defined by utilizing the regularization parameter as a variable based on the least squares solution of the minimization problem. Secondly, the quotient function method (QFM is proposed to select the optimal regularization parameter based on the quadratic programming theory. For employing the QFM, the characteristics of the values of QF with respect to the different regularization parameters are taken into consideration. Finally, numerical and experimental examples are utilized to validate the performance of the QFM. Furthermore, the Generalized Cross-Validation (GCV method and the L-curve method are taken as the comparison methods. The results indicate that the proposed QFM is adaptive to different measuring points, noise levels, and types of dynamic load.
Simulating non-Newtonian flows with the moving particle semi-implicit method with an SPH kernel
International Nuclear Information System (INIS)
Xiang, Hao; Chen, Bin
2015-01-01
The moving particle semi-implicit (MPS) method and smoothed particle hydrodynamics (SPH) are commonly used mesh-free particle methods for free surface flows. The MPS method has superiority in incompressible flow simulation and simple programing. However, the crude kernel function is not accurate enough for the discretization of the divergence of the shear stress tensor by the particle inconsistency when the MPS method is extended to non-Newtonian flows. This paper presents an improved MPS method with an SPH kernel to simulate non-Newtonian flows. To improve the consistency of the partial derivative, the SPH cubic spline kernel and the Taylor series expansion are combined with the MPS method. This approach is suitable for all non-Newtonian fluids that can be described with τ = μ(|γ|) Δ (where τ is the shear stress tensor, μ is the viscosity, |γ| is the shear rate, and Δ is the strain tensor), e.g., the Casson and Cross fluids. Two examples are simulated including the Newtonian Poiseuille flow and container filling process of the Cross fluid. The results of Poiseuille flow are more accurate than the traditional MPS method, and different filling processes are obtained with good agreement with previous results, which verified the validation of the new algorithm. For the Cross fluid, the jet fracture length can be correlated with We 0.28 Fr 0.78 (We is the Weber number, Fr is the Froude number). (paper)
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
On multiple level-set regularization methods for inverse problems
International Nuclear Information System (INIS)
DeCezaro, A; Leitão, A; Tai, X-C
2009-01-01
We analyze a multiple level-set method for solving inverse problems with piecewise constant solutions. This method corresponds to an iterated Tikhonov method for a particular Tikhonov functional G α based on TV–H 1 penalization. We define generalized minimizers for our Tikhonov functional and establish an existence result. Moreover, we prove convergence and stability results of the proposed Tikhonov method. A multiple level-set algorithm is derived from the first-order optimality conditions for the Tikhonov functional G α , similarly as the iterated Tikhonov method. The proposed multiple level-set method is tested on an inverse potential problem. Numerical experiments show that the method is able to recover multiple objects as well as multiple contrast levels
Choi, Sae Il
2009-01-01
This study used simulation (a) to compare the kernel equating method to traditional equipercentile equating methods under the equivalent-groups (EG) design and the nonequivalent-groups with anchor test (NEAT) design and (b) to apply the parametric bootstrap method for estimating standard errors of equating. A two-parameter logistic item response…
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.
Least square regularized regression in sum space.
Xu, Yong-Li; Chen, Di-Rong; Li, Han-Xiong; Liu, Lu
2013-04-01
This paper proposes a least square regularized regression algorithm in sum space of reproducing kernel Hilbert spaces (RKHSs) for nonflat function approximation, and obtains the solution of the algorithm by solving a system of linear equations. This algorithm can approximate the low- and high-frequency component of the target function with large and small scale kernels, respectively. The convergence and learning rate are analyzed. We measure the complexity of the sum space by its covering number and demonstrate that the covering number can be bounded by the product of the covering numbers of basic RKHSs. For sum space of RKHSs with Gaussian kernels, by choosing appropriate parameters, we tradeoff the sample error and regularization error, and obtain a polynomial learning rate, which is better than that in any single RKHS. The utility of this method is illustrated with two simulated data sets and five real-life databases.
Higher-Order Hybrid Gaussian Kernel in Meshsize Boosting Algorithm
African Journals Online (AJOL)
In this paper, we shall use higher-order hybrid Gaussian kernel in a meshsize boosting algorithm in kernel density estimation. Bias reduction is guaranteed in this scheme like other existing schemes but uses the higher-order hybrid Gaussian kernel instead of the regular fixed kernels. A numerical verification of this scheme ...
Adaptive Kernel in Meshsize Boosting Algorithm in KDE ...
African Journals Online (AJOL)
This paper proposes the use of adaptive kernel in a meshsize boosting algorithm in kernel density estimation. The algorithm is a bias reduction scheme like other existing schemes but uses adaptive kernel instead of the regular fixed kernels. An empirical study for this scheme is conducted and the findings are comparatively ...
Adaptive Kernel In The Bootstrap Boosting Algorithm In KDE ...
African Journals Online (AJOL)
This paper proposes the use of adaptive kernel in a bootstrap boosting algorithm in kernel density estimation. The algorithm is a bias reduction scheme like other existing schemes but uses adaptive kernel instead of the regular fixed kernels. An empirical study for this scheme is conducted and the findings are comparatively ...
Relaxation Methods for Strictly Convex Regularizations of Piecewise Linear Programs
International Nuclear Information System (INIS)
Kiwiel, K. C.
1998-01-01
We give an algorithm for minimizing the sum of a strictly convex function and a convex piecewise linear function. It extends several dual coordinate ascent methods for large-scale linearly constrained problems that occur in entropy maximization, quadratic programming, and network flows. In particular, it may solve exact penalty versions of such (possibly inconsistent) problems, and subproblems of bundle methods for nondifferentiable optimization. It is simple, can exploit sparsity, and in certain cases is highly parallelizable. Its global convergence is established in the recent framework of B -functions (generalized Bregman functions)
Smoothing-Norm Preconditioning for Regularizing Minimum-Residual Methods
DEFF Research Database (Denmark)
Hansen, Per Christian; Jensen, Toke Koldborg
2006-01-01
take into account a smoothing norm for the solution. This technique is well established for CGLS, but it does not immediately carry over to minimum-residual methods when the smoothing norm is a seminorm or a Sobolev norm. We develop a new technique which works for any smoothing norm of the form $\\|L...
Adaptive L1/2 Shooting Regularization Method for Survival Analysis Using Gene Expression Data
Directory of Open Access Journals (Sweden)
Xiao-Ying Liu
2013-01-01
Full Text Available A new adaptive L1/2 shooting regularization method for variable selection based on the Cox’s proportional hazards mode being proposed. This adaptive L1/2 shooting algorithm can be easily obtained by the optimization of a reweighed iterative series of L1 penalties and a shooting strategy of L1/2 penalty. Simulation results based on high dimensional artificial data show that the adaptive L1/2 shooting regularization method can be more accurate for variable selection than Lasso and adaptive Lasso methods. The results from real gene expression dataset (DLBCL also indicate that the L1/2 regularization method performs competitively.
Regularization parameter selection methods for ill-posed Poisson maximum likelihood estimation
International Nuclear Information System (INIS)
Bardsley, Johnathan M; Goldes, John
2009-01-01
In image processing applications, image intensity is often measured via the counting of incident photons emitted by the object of interest. In such cases, image data noise is accurately modeled by a Poisson distribution. This motivates the use of Poisson maximum likelihood estimation for image reconstruction. However, when the underlying model equation is ill-posed, regularization is needed. Regularized Poisson likelihood estimation has been studied extensively by the authors, though a problem of high importance remains: the choice of the regularization parameter. We will present three statistically motivated methods for choosing the regularization parameter, and numerical examples will be presented to illustrate their effectiveness
Study on preparation method of Zanthoxylum bungeanum seeds kernel oil with zero trans-fatty acids.
Liu, Tong; Yao, Shi-Yong; Yin, Zhong-Yi; Zheng, Xu-Xu; Shen, Yu
2016-04-01
The seed of Zanthoxylum bungeanum (Z. bungeanum) is a by-product of pepper production and rich in unsaturated fatty acid, cellulose, and protein. The seed oil obtained from traditional producing process by squeezing or extracting would be bad quality and could not be used as edible oil. In this paper, a new preparation method of Z. bungeanum seed kernel oil (ZSKO) was developed by comparing the advantages and disadvantages of alkali saponification-cold squeezing, alkali saponification-solvent extraction, and alkali saponification-supercritical fluid extraction with carbon dioxide (SFE-CO2). The results showed that the alkali saponification-cold squeezing could be the optimal preparation method of ZSKO, which contained the following steps: Z. bungeanum seed was pretreated by alkali saponification under the conditions of adding 10 %NaOH (w/w), solution temperature was 80 °C, and saponification reaction time was 45 min, and pretreated seed was separated by filtering, water washing, and overnight drying at 50 °C, then repeated squeezing was taken until no oil generated at 60 °C with 15 % moisture content, and ZSKO was attained finally using centrifuge. The produced ZSKO contained more than 90 % unsaturated fatty acids and no trans-fatty acids and be testified as a good edible oil with low-value level of acid and peroxide. It was demonstrated that the alkali saponification-cold squeezing process could be scaled up and applied to industrialized production of ZSKO.
Lan, C. E.; Lamar, J. E.
1977-01-01
A logarithmic-singularity correction factor is derived for use in kernel function methods associated with Multhopp's subsonic lifting-surface theory. Because of the form of the factor, a relation was formulated between the numbers of chordwise and spanwise control points needed for good accuracy. This formulation is developed and discussed. Numerical results are given to show the improvement of the computation with the new correction factor.
Regularization methods for inferential sensing in nuclear power plants
International Nuclear Information System (INIS)
Hines, J.W.; Gribok, A.V.; Attieh, I.; Uhrig, R.E.
2000-01-01
Inferential sensing is the use of information related to a plant parameter to infer its actual value. The most common method of inferential sensing uses a mathematical model to infer a parameter value from correlated sensor values. Collinearity in the predictor variables leads to an ill-posed problem that causes inconsistent results when data based models such as linear regression and neural networks are used. This chapter presents several linear and non-linear inferential sensing methods including linear regression and neural networks. Both of these methods can be modified from their original form to solve ill-posed problems and produce more consistent results. We will compare these techniques using data from Florida Power Corporation's Crystal River Nuclear Power Plant to predict the drift in a feedwater flow sensor. According to a report entitled 'Feedwater Flow Measurement in U.S. Nuclear Power Generation Stations' that was commissioned by the Electric Power Research Institute, venturi meter fouling is 'the single most frequent cause' for derating in Pressurized Water Reactors. This chapter presents several viable solutions to this problem. (orig.)
Energy Technology Data Exchange (ETDEWEB)
Li, L; Tan, S [Huazhong University of Science and Technology, Wuhan, Hubei (China); Lu, W [University of Maryland School of Medicine, Baltimore, MD (United States)
2015-06-15
Purpose: To propose a new variational method which couples image restoration with tumor segmentation for PET images using multiple regularizations. Methods: Partial volume effect (PVE) is a major degrading factor impacting tumor segmentation accuracy in PET imaging. The existing segmentation methods usually need to take prior calibrations to compensate PVE and they are highly system-dependent. Taking into account that image restoration and segmentation can promote each other and they are tightly coupled, we proposed a variational method to solve the two problems together. Our method integrated total variation (TV) semi-blind deconvolution and Mumford-Shah (MS) segmentation. The TV norm was used on edges to protect the edge information, and the L{sub 2} norm was used to avoid staircase effect in the no-edge area. The blur kernel was constrained to the Gaussian model parameterized by its variance and we assumed that the variances in the X-Y and Z directions are different. The energy functional was iteratively optimized by an alternate minimization algorithm. Segmentation performance was tested on eleven patients with non-Hodgkin’s lymphoma, and evaluated by Dice similarity index (DSI) and classification error (CE). For comparison, seven other widely used methods were also tested and evaluated. Results: The combination of TV and L{sub 2} regularizations effectively improved the segmentation accuracy. The average DSI increased by around 0.1 than using either the TV or the L{sub 2} norm. The proposed method was obviously superior to other tested methods. It has an average DSI and CE of 0.80 and 0.41, while the FCM method — the second best one — has only an average DSI and CE of 0.66 and 0.64. Conclusion: Coupling image restoration and segmentation can handle PVE and thus improves tumor segmentation accuracy in PET. Alternate use of TV and L2 regularizations can further improve the performance of the algorithm. This work was supported in part by National Natural
Online analysis of coal on a conveyor belt by use of machine vision and kernel methods
Energy Technology Data Exchange (ETDEWEB)
Aldrich, C.; Jemwa, G.T.; van Dyk, J.C.; Keyser, M.J.; van Heerden, J.H.P. [University of Stellenbosch, Stellenbosch (South Africa). Dept. of Process Engineering
2010-07-01
The objective of this project is to explore the use of image analysis to quantify the amount of fines (6mm) present for different coal samples under conditions simulating the coal on conveyor belts similar to those being used by Sasol for gasification purposes. Quantification of the fines will be deemed particularly successful, if the fines mass fraction, as determined by sieve analysis, is possible to be predicted with an error of less than 10%. In this article, kernel-based methods to estimate particle size ranges on a pilot-scale conveyor belt as well as edge detection algorithms are considered. Preliminary results have shown that the fines fraction in the coal on the conveyor belt could be estimated with a median error of approximately 24.1%. This analysis was based on a relatively small number of sieve samples (18 in total) and needs to be validated by more samples. More samples would also facilitate better calibration and may lead to improved estimates of the sieve fines fractions. Similarly, better results may also be possible by using different approaches to image acquisition and analysis. Most of the error in the fines estimates can be attributed to sampling and to fines that were randomly obscured by the top layer (of larger particles) of coal on the belt. Sampling errors occurred as a result of some breakage of the coal between the sieve analyses and the acquisition of the images. The percentage of the fines obscured by the top layer of the coal probably caused most of the variation in the estimated mass of fines, but this needs to be validated experimentally. Preliminary studies have indicated that some variation in the lighting conditions have a small influence on the reliability of the estimates of the coal fines fractions and that consistent lighting conditions are more important than optimal lighting conditions.
A kernel version of spatial factor analysis
DEFF Research Database (Denmark)
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...
Directory of Open Access Journals (Sweden)
Haotao Cai
2017-01-01
Full Text Available We develop a generalized Jacobi-Galerkin method for second kind Volterra integral equations with weakly singular kernels. In this method, we first introduce some known singular nonpolynomial functions in the approximation space of the conventional Jacobi-Galerkin method. Secondly, we use the Gauss-Jacobi quadrature rules to approximate the integral term in the resulting equation so as to obtain high-order accuracy for the approximation. Then, we establish that the approximate equation has a unique solution and the approximate solution arrives at an optimal convergence order. One numerical example is presented to demonstrate the effectiveness of the proposed method.
International Nuclear Information System (INIS)
Kowsary, F.; Pooladvand, K.; Pourshaghaghy, A.
2007-01-01
In this paper, an appropriate distribution of the heating elements' strengths in a radiation furnace is estimated using inverse methods so that a pre-specified temperature and heat flux distribution is attained on the design surface. Minimization of the sum of the squares of the error function is performed using the variable metric method (VMM), and the results are compared with those obtained by the conjugate gradient method (CGM) established previously in the literature. It is shown via test cases and a well-founded validation procedure that the VMM, when using a 'regularized' estimator, is more accurate and is able to reach at a higher quality final solution as compared to the CGM. The test cases used in this study were two-dimensional furnaces filled with an absorbing, emitting, and scattering gas
Directory of Open Access Journals (Sweden)
Jinping Tang
2017-01-01
Full Text Available Optical tomography is an emerging and important molecular imaging modality. The aim of optical tomography is to reconstruct optical properties of human tissues. In this paper, we focus on reconstructing the absorption coefficient based on the radiative transfer equation (RTE. It is an ill-posed parameter identification problem. Regularization methods have been broadly applied to reconstruct the optical coefficients, such as the total variation (TV regularization and the L1 regularization. In order to better reconstruct the piecewise constant and sparse coefficient distributions, TV and L1 norms are combined as the regularization. The forward problem is discretized with the discontinuous Galerkin method on the spatial space and the finite element method on the angular space. The minimization problem is solved by a Jacobian-based Levenberg-Marquardt type method which is equipped with a split Bregman algorithms for the L1 regularization. We use the adjoint method to compute the Jacobian matrix which dramatically improves the computation efficiency. By comparing with the other imaging reconstruction methods based on TV and L1 regularizations, the simulation results show the validity and efficiency of the proposed method.
A New Method for Determining Optimal Regularization Parameter in Near-Field Acoustic Holography
Directory of Open Access Journals (Sweden)
Yue Xiao
2018-01-01
Full Text Available Tikhonov regularization method is effective in stabilizing reconstruction process of the near-field acoustic holography (NAH based on the equivalent source method (ESM, and the selection of the optimal regularization parameter is a key problem that determines the regularization effect. In this work, a new method for determining the optimal regularization parameter is proposed. The transfer matrix relating the source strengths of the equivalent sources to the measured pressures on the hologram surface is augmented by adding a fictitious point source with zero strength. The minimization of the norm of this fictitious point source strength is as the criterion for choosing the optimal regularization parameter since the reconstructed value should tend to zero. The original inverse problem in calculating the source strengths is converted into a univariate optimization problem which is solved by a one-dimensional search technique. Two numerical simulations with a point driven simply supported plate and a pulsating sphere are investigated to validate the performance of the proposed method by comparison with the L-curve method. The results demonstrate that the proposed method can determine the regularization parameter correctly and effectively for the reconstruction in NAH.
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
Quantum tomography, phase-space observables and generalized Markov kernels
International Nuclear Information System (INIS)
Pellonpaeae, Juha-Pekka
2009-01-01
We construct a generalized Markov kernel which transforms the observable associated with the homodyne tomography into a covariant phase-space observable with a regular kernel state. Illustrative examples are given in the cases of a 'Schroedinger cat' kernel state and the Cahill-Glauber s-parametrized distributions. Also we consider an example of a kernel state when the generalized Markov kernel cannot be constructed.
International Nuclear Information System (INIS)
Cao, A.
1981-07-01
This study is concerned with the transverse axial gamma emission tomography. The problem of self-attenuation of radiations in biologic tissues is raised. The regularizing iterative method is developed, as a reconstruction method of 3 dimensional images. The different steps from acquisition to results, necessary to its application, are described. Organigrams relative to each step are explained. Comparison notion between two reconstruction methods is introduced. Some methods used for the comparison or to bring about the characteristics of a reconstruction technique are defined. The studies realized to test the regularizing iterative method are presented and results are analyzed [fr
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
Burman, Erik; Hansbo, Peter; Larson, Mats G.
2018-03-01
Tikhonov regularization is one of the most commonly used methods for the regularization of ill-posed problems. In the setting of finite element solutions of elliptic partial differential control problems, Tikhonov regularization amounts to adding suitably weighted least squares terms of the control variable, or derivatives thereof, to the Lagrangian determining the optimality system. In this note we show that the stabilization methods for discretely ill-posed problems developed in the setting of convection-dominated convection-diffusion problems, can be highly suitable for stabilizing optimal control problems, and that Tikhonov regularization will lead to less accurate discrete solutions. We consider some inverse problems for Poisson’s equation as an illustration and derive new error estimates both for the reconstruction of the solution from the measured data and reconstruction of the source term from the measured data. These estimates include both the effect of the discretization error and error in the measurements.
International Nuclear Information System (INIS)
Li, Boyan; Calvet, Amandine; Casamayou-Boucau, Yannick; Ryder, Alan G.
2016-01-01
A new, fully automated, rapid method, referred to as kernel principal component analysis residual diagnosis (KPCARD), is proposed for removing cosmic ray artifacts (CRAs) in Raman spectra, and in particular for large Raman imaging datasets. KPCARD identifies CRAs via a statistical analysis of the residuals obtained at each wavenumber in the spectra. The method utilizes the stochastic nature of CRAs; therefore, the most significant components in principal component analysis (PCA) of large numbers of Raman spectra should not contain any CRAs. The process worked by first implementing kernel PCA (kPCA) on all the Raman mapping data and second accurately estimating the inter- and intra-spectrum noise to generate two threshold values. CRA identification was then achieved by using the threshold values to evaluate the residuals for each spectrum and assess if a CRA was present. CRA correction was achieved by spectral replacement where, the nearest neighbor (NN) spectrum, most spectroscopically similar to the CRA contaminated spectrum and principal components (PCs) obtained by kPCA were both used to generate a robust, best curve fit to the CRA contaminated spectrum. This best fit spectrum then replaced the CRA contaminated spectrum in the dataset. KPCARD efficacy was demonstrated by using simulated data and real Raman spectra collected from solid-state materials. The results showed that KPCARD was fast ( 1 million) Raman datasets. - Highlights: • New rapid, automatable method for cosmic ray artifact correction of Raman spectra. • Uses combination of kernel PCA and noise estimation for artifact identification. • Implements a best fit spectrum replacement correction approach.
Directory of Open Access Journals (Sweden)
Teguh Prawono Sabat
2017-08-01
Full Text Available Geographically Weighted Logistic Regression (GWLR was regression model consider the spatial factor, which could be used to analyze the IMR. The number of Infant Mortality as big as 100 cases in 2015 or 12 per 1000 live birth in South Central Timor Regency. The aim of this study was to determine the best modeling of GWLR with fixed weighting function and Adaptive Gaussian Kernel in the case of infant mortality in South Central Timor District in 2015. The response variable (Y in this study was a case of infant mortality, while variable predictor was the percentage of neonatal first visit (KN1 (X1, the percentage of neonatal visit 3 times (Complete KN (X2, the percentage of pregnant get Fe tablet (X3, percentage of poor families pre prosperous (X4. This was a non-reactive study, which is a measurement which individuals surveyed did not realize that they are part of a study, with analysis unit in 32 sub-districts of South Central Timor Districts. Data analysis used open source program that was Excel, R program, Quantum GIS and GWR4. The best GWLR spatial modeling with Adaptive Gaussian Kernel weighting function, a global model parameters GWLR Adaptive Gaussian Kernel weighting function obtained by g (x = 0.941086 - 0,892506X4, GWLR local models with adaptive Kernel bisquare weighting function in the 13 Districts were obtained g(x = 0 − 0X4, factors that affect the cases of infant mortality in 13 sub-districts of South Central Timor Regency in 2015 was the percentage of poor families pre prosperous.
Analysis of the iteratively regularized Gauss-Newton method under a heuristic rule
Jin, Qinian; Wang, Wei
2018-03-01
The iteratively regularized Gauss-Newton method is one of the most prominent regularization methods for solving nonlinear ill-posed inverse problems when the data is corrupted by noise. In order to produce a useful approximate solution, this iterative method should be terminated properly. The existing a priori and a posteriori stopping rules require accurate information on the noise level, which may not be available or reliable in practical applications. In this paper we propose a heuristic selection rule for this regularization method, which requires no information on the noise level. By imposing certain conditions on the noise, we derive a posteriori error estimates on the approximate solutions under various source conditions. Furthermore, we establish a convergence result without using any source condition. Numerical results are presented to illustrate the performance of our heuristic selection rule.
Theory of reproducing kernels and applications
Saitoh, Saburou
2016-01-01
This book provides a large extension of the general theory of reproducing kernels published by N. Aronszajn in 1950, with many concrete applications. In Chapter 1, many concrete reproducing kernels are first introduced with detailed information. Chapter 2 presents a general and global theory of reproducing kernels with basic applications in a self-contained way. Many fundamental operations among reproducing kernel Hilbert spaces are dealt with. Chapter 2 is the heart of this book. Chapter 3 is devoted to the Tikhonov regularization using the theory of reproducing kernels with applications to numerical and practical solutions of bounded linear operator equations. In Chapter 4, the numerical real inversion formulas of the Laplace transform are presented by applying the Tikhonov regularization, where the reproducing kernels play a key role in the results. Chapter 5 deals with ordinary differential equations; Chapter 6 includes many concrete results for various fundamental partial differential equations. In Chapt...
Directory of Open Access Journals (Sweden)
Fairouz Zouyed
2015-01-01
Full Text Available This paper discusses the inverse problem of determining an unknown source in a second order differential equation from measured final data. This problem is ill-posed; that is, the solution (if it exists does not depend continuously on the data. In order to solve the considered problem, an iterative method is proposed. Using this method a regularized solution is constructed and an a priori error estimate between the exact solution and its regularized approximation is obtained. Moreover, numerical results are presented to illustrate the accuracy and efficiency of this method.
Iterative Method of Regularization with Application of Advanced Technique for Detection of Contours
International Nuclear Information System (INIS)
Niedziela, T.; Stankiewicz, A.
2000-01-01
This paper proposes a novel iterative method of regularization with application of an advanced technique for detection of contours. To eliminate noises, the properties of convolution of functions are utilized. The method can be accomplished in a simple neural cellular network, which creates the possibility of extraction of contours by automatic image recognition equipment. (author)
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
Directory of Open Access Journals (Sweden)
S. Ceccherini
2007-01-01
Full Text Available The retrieval of concentration vertical profiles of atmospheric constituents from spectroscopic measurements is often an ill-conditioned problem and regularization methods are frequently used to improve its stability. Recently a new method, that provides a good compromise between precision and vertical resolution, was proposed to determine analytically the value of the regularization parameter. This method is applied for the first time to real measurements with its implementation in the operational retrieval code of the satellite limb-emission measurements of the MIPAS instrument and its performances are quantitatively analyzed. The adopted regularization improves the stability of the retrieval providing smooth profiles without major degradation of the vertical resolution. In the analyzed measurements the retrieval procedure provides a vertical resolution that, in the troposphere and low stratosphere, is smaller than the vertical field of view of the instrument.
Global Polynomial Kernel Hazard Estimation
DEFF Research Database (Denmark)
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...
Local Observed-Score Kernel Equating
Wiberg, Marie; van der Linden, Wim J.; von Davier, Alina A.
2014-01-01
Three local observed-score kernel equating methods that integrate methods from the local equating and kernel equating frameworks are proposed. The new methods were compared with their earlier counterparts with respect to such measures as bias--as defined by Lord's criterion of equity--and percent relative error. The local kernel item response…
Use of regularization method in the determination of ring parameters and orbit correction
International Nuclear Information System (INIS)
Tang, Y.N.; Krinsky, S.
1993-01-01
We discuss applying the regularization method of Tikhonov to the solution of inverse problems arising in accelerator operations. This approach has been successfully used for orbit correction on the NSLS storage rings, and is presently being applied to the determination of betatron functions and phases from the measured response matrix. The inverse problem of differential equation often leads to a set of integral equations of the first kind which are ill-conditioned. The regularization method is used to combat the ill-posedness
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.
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.
Information operator approach and iterative regularization methods for atmospheric remote sensing
International Nuclear Information System (INIS)
Doicu, A.; Hilgers, S.; Bargen, A. von; Rozanov, A.; Eichmann, K.-U.; Savigny, C. von; Burrows, J.P.
2007-01-01
In this study, we present the main features of the information operator approach for solving linear inverse problems arising in atmospheric remote sensing. This method is superior to the stochastic version of the Tikhonov regularization (or the optimal estimation method) due to its capability to filter out the noise-dominated components of the solution generated by an inappropriate choice of the regularization parameter. We extend this approach to iterative methods for nonlinear ill-posed problems and derive the truncated versions of the Gauss-Newton and Levenberg-Marquardt methods. Although the paper mostly focuses on discussing the mathematical details of the inverse method, retrieval results have been provided, which exemplify the performances of the methods. These results correspond to the NO 2 retrieval from SCIAMACHY limb scatter measurements and have been obtained by using the retrieval processors developed at the German Aerospace Center Oberpfaffenhofen and Institute of Environmental Physics of the University of Bremen
International Nuclear Information System (INIS)
Akaho, E.H.K.; Dagadu, C.P.K.; Maaku, B.T.; Anim-Sampong, S.; Kyere, A.W.K.; Jonah, S.A.
2001-01-01
A fast and non-destructive technique based on thermal neutron moderation has been used for determining the total hydrogen content in two types of red palm oil (dzomi and amidze) and palm kernel oil produced by traditional methods in Ghana. An equipment consisting of an 241 Am-Be neutron source and 3 He neutron detector was used in the investigation. The equipment was originally designed for detection of liquid levels in petrochemical and other process industries. Standards in the form of liquid hydrocarbons were used to obtain calibration lines for thermal neutron reflection parameter as a function of hydrogen content. Measured reflection parameters with respective hydrogen content with or without heat treatment of the three edible palm oils available on the market were compared with a brand cooking oil (frytol). The average total hydrogen content in the local oil samples prior to heating was measured to be 11.62 w% which compared well with acceptable value of 12 w% for palm oils in the sub-region. After heat treatment, the frytol oil (produced through bleaching process) had the least loss of hydrogen content of 0.26% in comparison with palm kernel oil of 0.44% followed by dzomi of 1.96% and by amidze of 3.22%. (author)
Directory of Open Access Journals (Sweden)
Zhao-Qing Wang
2014-01-01
Full Text Available Embedding the irregular doubly connected domain into an annular regular region, the unknown functions can be approximated by the barycentric Lagrange interpolation in the regular region. A highly accurate regular domain collocation method is proposed for solving potential problems on the irregular doubly connected domain in polar coordinate system. The formulations of regular domain collocation method are constructed by using barycentric Lagrange interpolation collocation method on the regular domain in polar coordinate system. The boundary conditions are discretized by barycentric Lagrange interpolation within the regular domain. An additional method is used to impose the boundary conditions. The least square method can be used to solve the overconstrained equations. The function values of points in the irregular doubly connected domain can be calculated by barycentric Lagrange interpolation within the regular domain. Some numerical examples demonstrate the effectiveness and accuracy of the presented method.
A kernel version of multivariate alteration detection
DEFF Research Database (Denmark)
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....
A regularization method for solving the Poisson equation for mixed unbounded-periodic domains
DEFF Research Database (Denmark)
Spietz, Henrik Juul; Mølholm Hejlesen, Mads; Walther, Jens Honoré
2018-01-01
the regularized unbounded-periodic Green's functions can be implemented in an FFT-based Poisson solver to obtain a convergence rate corresponding to the regularization order of the Green's function. The high order is achieved without any additional computational cost from the conventional FFT-based Poisson solver...... and enables the calculation of the derivative of the solution to the same high order by direct spectral differentiation. We illustrate an application of the FFT-based Poisson solver by using it with a vortex particle mesh method for the approximation of incompressible flow for a problem with a single periodic...
Regularization methods for ill-posed problems in multiple Hilbert scales
International Nuclear Information System (INIS)
Mazzieri, Gisela L; Spies, Ruben D
2012-01-01
Several convergence results in Hilbert scales under different source conditions are proved and orders of convergence and optimal orders of convergence are derived. Also, relations between those source conditions are proved. The concept of a multiple Hilbert scale on a product space is introduced, and regularization methods on these scales are defined, both for the case of a single observation and for the case of multiple observations. In the latter case, it is shown how vector-valued regularization functions in these multiple Hilbert scales can be used. In all cases, convergence is proved and orders and optimal orders of convergence are shown. Finally, some potential applications and open problems are discussed. (paper)
Laplacian embedded regression for scalable manifold regularization.
Chen, Lin; Tsang, Ivor W; Xu, Dong
2012-06-01
Semi-supervised learning (SSL), as a powerful tool to learn from a limited number of labeled data and a large number of unlabeled data, has been attracting increasing attention in the machine learning community. In particular, the manifold regularization framework has laid solid theoretical foundations for a large family of SSL algorithms, such as Laplacian support vector machine (LapSVM) and Laplacian regularized least squares (LapRLS). However, most of these algorithms are limited to small scale problems due to the high computational cost of the matrix inversion operation involved in the optimization problem. In this paper, we propose a novel framework called Laplacian embedded regression by introducing an intermediate decision variable into the manifold regularization framework. By using ∈-insensitive loss, we obtain the Laplacian embedded support vector regression (LapESVR) algorithm, which inherits the sparse solution from SVR. Also, we derive Laplacian embedded RLS (LapERLS) corresponding to RLS under the proposed framework. Both LapESVR and LapERLS possess a simpler form of a transformed kernel, which is the summation of the original kernel and a graph kernel that captures the manifold structure. The benefits of the transformed kernel are two-fold: (1) we can deal with the original kernel matrix and the graph Laplacian matrix in the graph kernel separately and (2) if the graph Laplacian matrix is sparse, we only need to perform the inverse operation for a sparse matrix, which is much more efficient when compared with that for a dense one. Inspired by kernel principal component analysis, we further propose to project the introduced decision variable into a subspace spanned by a few eigenvectors of the graph Laplacian matrix in order to better reflect the data manifold, as well as accelerate the calculation of the graph kernel, allowing our methods to efficiently and effectively cope with large scale SSL problems. Extensive experiments on both toy and real
New method for minimizing regular functions with constraints on parameter region
International Nuclear Information System (INIS)
Kurbatov, V.S.; Silin, I.N.
1993-01-01
The new method of function minimization is developed. Its main features are considered. It is possible minimization of regular function with the arbitrary structure. For χ 2 -like function the usage of simplified second derivatives is possible with the control of correctness. The constraints of arbitrary structure can be used. The means for fast movement along multidimensional valleys are used. The method is tested on real data of K π2 decay of the experiment on rare K - -decays. 6 refs
On Landweber–Kaczmarz methods for regularizing systems of ill-posed equations in Banach spaces
International Nuclear Information System (INIS)
Leitão, A; Alves, M Marques
2012-01-01
In this paper, iterative regularization methods of Landweber–Kaczmarz type are considered for solving systems of ill-posed equations modeled (finitely many) by operators acting between Banach spaces. Using assumptions of uniform convexity and smoothness on the parameter space, we are able to prove a monotony result for the proposed method, as well as to establish convergence (for exact data) and stability results (in the noisy data case). (paper)
International Nuclear Information System (INIS)
Xu, Yanbin; Pei, Yang; Dong, Feng
2016-01-01
The L-curve method is a popular regularization parameter choice method for the ill-posed inverse problem of electrical resistance tomography (ERT). However the method cannot always determine a proper parameter for all situations. An investigation into those situations where the L-curve method failed show that a new corner point appears on the L-curve and the parameter corresponding to the new corner point can obtain a satisfactory reconstructed solution. Thus an extended L-curve method, which determines the regularization parameter associated with either global corner or the new corner, is proposed. Furthermore, two strategies are provided to determine the new corner–one is based on the second-order differential of L-curve, and the other is based on the curvature of L-curve. The proposed method is examined by both numerical simulations and experimental tests. And the results indicate that the extended method can handle the parameter choice problem even in the case where the typical L-curve method fails. Finally, in order to reduce the running time of the method, the extended method is combined with a projection method based on the Krylov subspace, which was able to boost the extended L-curve method. The results verify that the speed of the extended L-curve method is distinctly improved. The proposed method extends the application of the L-curve in the field of choosing regularization parameter with an acceptable running time and can also be used in other kinds of tomography. (paper)
International Nuclear Information System (INIS)
Huang, J; Followill, D; Howell, R; Liu, X; Mirkovic, D; Stingo, F; Kry, S
2015-01-01
Purpose: To investigate two strategies for reducing dose calculation errors near metal implants: use of CT metal artifact reduction methods and implementation of metal-based energy deposition kernels in the convolution/superposition (C/S) method. Methods: Radiochromic film was used to measure the dose upstream and downstream of titanium and Cerrobend implants. To assess the dosimetric impact of metal artifact reduction methods, dose calculations were performed using baseline, uncorrected images and metal artifact reduction Methods: Philips O-MAR, GE’s monochromatic gemstone spectral imaging (GSI) using dual-energy CT, and GSI imaging with metal artifact reduction software applied (MARs).To assess the impact of metal kernels, titanium and silver kernels were implemented into a commercial collapsed cone C/S algorithm. Results: The CT artifact reduction methods were more successful for titanium than Cerrobend. Interestingly, for beams traversing the metal implant, we found that errors in the dimensions of the metal in the CT images were more important for dose calculation accuracy than reduction of imaging artifacts. The MARs algorithm caused a distortion in the shape of the titanium implant that substantially worsened the calculation accuracy. In comparison to water kernel dose calculations, metal kernels resulted in better modeling of the increased backscatter dose at the upstream interface but decreased accuracy directly downstream of the metal. We also found that the success of metal kernels was dependent on dose grid size, with smaller calculation voxels giving better accuracy. Conclusion: Our study yielded mixed results, with neither the metal artifact reduction methods nor the metal kernels being globally effective at improving dose calculation accuracy. However, some successes were observed. The MARs algorithm decreased errors downstream of Cerrobend by a factor of two, and metal kernels resulted in more accurate backscatter dose upstream of metals. Thus
A self-adapting and altitude-dependent regularization method for atmospheric profile retrievals
Directory of Open Access Journals (Sweden)
M. Ridolfi
2009-03-01
Full Text Available MIPAS is a Fourier transform spectrometer, operating onboard of the ENVISAT satellite since July 2002. The online retrieval algorithm produces geolocated profiles of temperature and of volume mixing ratios of six key atmospheric constituents: H_{2}O, O_{3}, HNO_{3}, CH_{4}, N_{2}O and NO_{2}. In the validation phase, oscillations beyond the error bars were observed in several profiles, particularly in CH_{4} and N_{2}O.
To tackle this problem, a Tikhonov regularization scheme has been implemented in the retrieval algorithm. The applied regularization is however rather weak in order to preserve the vertical resolution of the profiles.
In this paper we present a self-adapting and altitude-dependent regularization approach that detects whether the analyzed observations contain information about small-scale profile features, and determines the strength of the regularization accordingly. The objective of the method is to smooth out artificial oscillations as much as possible, while preserving the fine detail features of the profile when related information is detected in the observations.
The proposed method is checked for self consistency, its performance is tested on MIPAS observations and compared with that of some other regularization schemes available in the literature. In all the considered cases the proposed scheme achieves a good performance, thanks to its altitude dependence and to the constraints employed, which are specific of the inversion problem under consideration. The proposed method is generally applicable to iterative Gauss-Newton algorithms for the retrieval of vertical distribution profiles from atmospheric remote sounding measurements.
Output regularization of SVM seizure predictors: Kalman Filter versus the "Firing Power" method.
Teixeira, Cesar; Direito, Bruno; Bandarabadi, Mojtaba; Dourado, António
2012-01-01
Two methods for output regularization of support vector machines (SVMs) classifiers were applied for seizure prediction in 10 patients with long-term annotated data. The output of the classifiers were regularized by two methods: one based on the Kalman Filter (KF) and other based on a measure called the "Firing Power" (FP). The FP is a quantification of the rate of the classification in the preictal class in a past time window. In order to enable the application of the KF, the classification problem was subdivided in a two two-class problem, and the real-valued output of SVMs was considered. The results point that the FP method raise less false alarms than the KF approach. However, the KF approach presents an higher sensitivity, but the high number of false alarms turns their applicability negligible in some situations.
International Nuclear Information System (INIS)
Gupta, N.K.
1981-01-01
A new coupling kernel is developed for the three-dimensional (3-D) simulation of Boiling Water Reactors (BWR's) by the nodal coupling method. The new kernel depends not only on the properties of the node under consideration but also on the properties of its neighbouring nodes. This makes the kernel more useful in particular for fuel bundles lying in a surrounding of different nuclear characteristics, e.g. for a controlled bundle in the surrounding of uncontrolled bundles or vice-versa. The main parameter in the new kernel is a space-dependent factor obtained from the ratio of thermal-to-fast flux. The average value of the above ratio for each node is evaluated analytically. The kernel is incorporated in a 3-D BWR core simulation program MOGS. As an experimental verification of the model, the cycle-6 operations of the two units of the Tarapur Atomic Power Station (TAPS) are simulated and the result of the simulation are compared with Travelling Incore Probe (TIP) data. (orig.)
Directory of Open Access Journals (Sweden)
Liquan Mei
2014-01-01
Full Text Available A Galerkin method for a modified regularized long wave equation is studied using finite elements in space, the Crank-Nicolson scheme, and the Runge-Kutta scheme in time. In addition, an extrapolation technique is used to transform a nonlinear system into a linear system in order to improve the time accuracy of this method. A Fourier stability analysis for the method is shown to be marginally stable. Three invariants of motion are investigated. Numerical experiments are presented to check the theoretical study of this method.
Kulkarni, Ankur H; Ghosh, Prasenjit; Seetharaman, Ashwin; Kondaiah, Paturu; Gundiah, Namrata
2018-05-09
Traction forces exerted by adherent cells are quantified using displacements of embedded markers on polyacrylamide substrates due to cell contractility. Fourier Transform Traction Cytometry (FTTC) is widely used to calculate tractions but has inherent limitations due to errors in the displacement fields; these are mitigated through a regularization parameter (γ) in the Reg-FTTC method. An alternate finite element (FE) approach computes tractions on a domain using known boundary conditions. Robust verification and recovery studies are lacking but essential in assessing the accuracy and noise sensitivity of the traction solutions from the different methods. We implemented the L2 regularization method and defined a maximum curvature point in the traction with γ plot as the optimal regularization parameter (γ*) in the Reg-FTTC approach. Traction reconstructions using γ* yield accurate values of low and maximum tractions (Tmax) in the presence of up to 5% noise. Reg-FTTC is hence a clear improvement over the FTTC method but is inadequate to reconstruct low stresses such as those at nascent focal adhesions. FE, implemented using a node-by-node comparison, showed an intermediate reconstruction compared to Reg-FTTC. We performed experiments using mouse embryonic fibroblast (MEF) and compared results between these approaches. Tractions from FTTC and FE showed differences of ∼92% and 22% as compared to Reg-FTTC. Selection of an optimum value of γ for each cell reduced variability in the computed tractions as compared to using a single value of γ for all the MEF cells in this study.
Energy Technology Data Exchange (ETDEWEB)
Chen, Xueli, E-mail: xlchen@xidian.edu.cn, E-mail: jimleung@mail.xidian.edu.cn; Yang, Defu; Zhang, Qitan; Liang, Jimin, E-mail: xlchen@xidian.edu.cn, E-mail: jimleung@mail.xidian.edu.cn [School of Life Science and Technology, Xidian University, Xi' an 710071 (China); Engineering Research Center of Molecular and Neuro Imaging, Ministry of Education (China)
2014-05-14
Even though bioluminescence tomography (BLT) exhibits significant potential and wide applications in macroscopic imaging of small animals in vivo, the inverse reconstruction is still a tough problem that has plagued researchers in a related area. The ill-posedness of inverse reconstruction arises from insufficient measurements and modeling errors, so that the inverse reconstruction cannot be solved directly. In this study, an l{sub 1/2} regularization based numerical method was developed for effective reconstruction of BLT. In the method, the inverse reconstruction of BLT was constrained into an l{sub 1/2} regularization problem, and then the weighted interior-point algorithm (WIPA) was applied to solve the problem through transforming it into obtaining the solution of a series of l{sub 1} regularizers. The feasibility and effectiveness of the proposed method were demonstrated with numerical simulations on a digital mouse. Stability verification experiments further illustrated the robustness of the proposed method for different levels of Gaussian noise.
Regularization of DT-MRI Using 3D Median Filtering Methods
Directory of Open Access Journals (Sweden)
Soondong Kwon
2014-01-01
Full Text Available DT-MRI (diffusion tensor magnetic resonance imaging tractography is a method to determine the architecture of axonal fibers in the central nervous system by computing the direction of the principal eigenvectors obtained from tensor matrix, which is different from the conventional isotropic MRI. Tractography based on DT-MRI is known to need many computations and is highly sensitive to noise. Hence, adequate regularization methods, such as image processing techniques, are in demand. Among many regularization methods we are interested in the median filtering method. In this paper, we extended two-dimensional median filters already developed to three-dimensional median filters. We compared four median filtering methods which are two-dimensional simple median method (SM2D, two-dimensional successive Fermat method (SF2D, three-dimensional simple median method (SM3D, and three-dimensional successive Fermat method (SF3D. Three kinds of synthetic data with different altitude angles from axial slices and one kind of human data from MR scanner are considered for numerical implementation by the four filtering methods.
Model selection in kernel ridge regression
DEFF Research Database (Denmark)
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...
Phase reconstruction by a multilevel iteratively regularized Gauss–Newton method
International Nuclear Information System (INIS)
Langemann, Dirk; Tasche, Manfred
2008-01-01
In this paper we consider the numerical solution of a phase retrieval problem for a compactly supported, linear spline f : R → C with the Fourier transform f-circumflex, where values of |f| and |f-circumflex| at finitely many equispaced nodes are given. The unknown phases of complex spline coefficients fulfil a well-structured system of nonlinear equations. Thus the phase reconstruction leads to a nonlinear inverse problem, which is solved by a multilevel strategy and iterative Tikhonov regularization. The multilevel strategy concentrates the main effort of the solution of the phase retrieval problem in the coarse, less expensive levels and provides convenient initial guesses at the next finer level. On each level, the corresponding nonlinear system is solved by an iteratively regularized Gauss–Newton method. The multilevel strategy is motivated by convergence results of IRGN. This method is applicable to a wide range of examples as shown in several numerical tests for noiseless and noisy data
2016-11-22
structure of the graph, we replace the ℓ1- norm by the nonconvex Capped -ℓ1 norm , and obtain the Generalized Capped -ℓ1 regularized logistic regression...X. M. Yuan. Linearized augmented lagrangian and alternating direction methods for nuclear norm minimization. Mathematics of Computation, 82(281):301...better approximations of ℓ0- norm theoretically and computationally beyond ℓ1- norm , for example, the compressive sensing (Xiao et al., 2011). The
On the evaluation of X-ray diffraction experiments by the regularization method
Energy Technology Data Exchange (ETDEWEB)
Trubin, V.A.; Szasz, A. (Lab. of Surface and Interface Physics, Eoetvoes Univ., Budapest (Hungary))
1991-05-16
The characteristic property of diffractometers as the presence of occasional and systematic errors in measured patterns requires such an evaluation which is as informative as possible. This circumstance gives rise to the problem of optimal planning of the experiment. The X-ray diffraction optimization problem with application of the regularization method is studied. The proposal permits to determine more accurately the unknown true characteristics of the X-ray diffraction experiment. (orig.).
On the evaluation of X-ray diffraction experiments by the regularization method
International Nuclear Information System (INIS)
Trubin, V.A.; Szasz, A.
1991-01-01
The characteristic property of diffractometers as the presence of occasional and systematic errors in measured patterns requires such an evaluation which is as informative as possible. This circumstance gives rise to the problem of optimal planning of the experiment. The X-ray diffraction optimization problem with application of the regularization method is studied. The proposal permits to determine more accurately the unknown true characteristics of the X-ray diffraction experiment. (orig.)
A regularized vortex-particle mesh method for large eddy simulation
DEFF Research Database (Denmark)
Spietz, Henrik Juul; Walther, Jens Honore; Hejlesen, Mads Mølholm
We present recent developments of the remeshed vortex particle-mesh method for simulating incompressible ﬂuid ﬂow. The presented method relies on a parallel higher-order FFT based solver for the Poisson equation. Arbitrary high order is achieved through regularization of singular Green’s function...... solutions to the Poisson equation and recently we have derived novel high order solutions for a mixture of open and periodic domains. With this approach the simulated variables may formally be viewed as the approximate solution to the ﬁltered Navier Stokes equations, hence we use the method for Large Eddy...
Fibonacci-regularization method for solving Cauchy integral equations of the first kind
Directory of Open Access Journals (Sweden)
Mohammad Ali Fariborzi Araghi
2017-09-01
Full Text Available In this paper, a novel scheme is proposed to solve the first kind Cauchy integral equation over a finite interval. For this purpose, the regularization method is considered. Then, the collocation method with Fibonacci base function is applied to solve the obtained second kind singular integral equation. Also, the error estimate of the proposed scheme is discussed. Finally, some sample Cauchy integral equations stem from the theory of airfoils in fluid mechanics are presented and solved to illustrate the importance and applicability of the given algorithm. The tables in the examples show the efficiency of the method.
Application of L1/2 regularization logistic method in heart disease diagnosis.
Zhang, Bowen; Chai, Hua; Yang, Ziyi; Liang, Yong; Chu, Gejin; Liu, Xiaoying
2014-01-01
Heart disease has become the number one killer of human health, and its diagnosis depends on many features, such as age, blood pressure, heart rate and other dozens of physiological indicators. Although there are so many risk factors, doctors usually diagnose the disease depending on their intuition and experience, which requires a lot of knowledge and experience for correct determination. To find the hidden medical information in the existing clinical data is a noticeable and powerful approach in the study of heart disease diagnosis. In this paper, sparse logistic regression method is introduced to detect the key risk factors using L(1/2) regularization on the real heart disease data. Experimental results show that the sparse logistic L(1/2) regularization method achieves fewer but informative key features than Lasso, SCAD, MCP and Elastic net regularization approaches. Simultaneously, the proposed method can cut down the computational complexity, save cost and time to undergo medical tests and checkups, reduce the number of attributes needed to be taken from patients.
Ma, Denglong; Tan, Wei; Zhang, Zaoxiao; Hu, Jun
2017-03-05
In order to identify the parameters of hazardous gas emission source in atmosphere with less previous information and reliable probability estimation, a hybrid algorithm coupling Tikhonov regularization with particle swarm optimization (PSO) was proposed. When the source location is known, the source strength can be estimated successfully by common Tikhonov regularization method, but it is invalid when the information about both source strength and location is absent. Therefore, a hybrid method combining linear Tikhonov regularization and PSO algorithm was designed. With this method, the nonlinear inverse dispersion model was transformed to a linear form under some assumptions, and the source parameters including source strength and location were identified simultaneously by linear Tikhonov-PSO regularization method. The regularization parameters were selected by L-curve method. The estimation results with different regularization matrixes showed that the confidence interval with high-order regularization matrix is narrower than that with zero-order regularization matrix. But the estimation results of different source parameters are close to each other with different regularization matrixes. A nonlinear Tikhonov-PSO hybrid regularization was also designed with primary nonlinear dispersion model to estimate the source parameters. The comparison results of simulation and experiment case showed that the linear Tikhonov-PSO method with transformed linear inverse model has higher computation efficiency than nonlinear Tikhonov-PSO method. The confidence intervals from linear Tikhonov-PSO are more reasonable than that from nonlinear method. The estimation results from linear Tikhonov-PSO method are similar to that from single PSO algorithm, and a reasonable confidence interval with some probability levels can be additionally given by Tikhonov-PSO method. Therefore, the presented linear Tikhonov-PSO regularization method is a good potential method for hazardous emission
Sasipriya, Gopalakrishnan; Siddhuraju, Perumal
2012-08-01
The present study is proposed to determine the antioxidant activity of raw and processed samples of underutilized legumes, Entada scandens seed kernel and Canavalia gladiata seeds. The indigenous processing methods like dry heating, autoclaving and soaking followed by autoclaving in different solutions (plain water, ash, sugar and sodium bicarbonate) were adopted to seed samples. All other processing methods than dry heat showed significant reduction in phenolics (2.9-63%), tannins (26-100%) and flavonoids (14-67%). However, in processed samples of E. scandens, the hydroxyl radical scavenging activity and β-carotene bleaching inhibition activity were increased, whereas, 2,2-azinobis (3-ethyl benzothiazoline-6-sulfonic acid) diammonium salt (ABTS·(+)), ferric reducing antioxidant power (FRAP), metal chelating and superoxide anion scavenging activity were similar to unprocessed ones. In contrary, except dry heating in C. gladiata, all other processing methods significantly (Pprocessing methods in E. scandens and dry heating in C. gladiata would be a suitable method for adopting in domestic or industrial processing. Copyright © 2012 Elsevier Ltd. All rights reserved.
Bai, Bing
2012-03-01
There has been a lot of work on total variation (TV) regularized tomographic image reconstruction recently. Many of them use gradient-based optimization algorithms with a differentiable approximation of the TV functional. In this paper we apply TV regularization in Positron Emission Tomography (PET) image reconstruction. We reconstruct the PET image in a Bayesian framework, using Poisson noise model and TV prior functional. The original optimization problem is transformed to an equivalent problem with inequality constraints by adding auxiliary variables. Then we use an interior point method with logarithmic barrier functions to solve the constrained optimization problem. In this method, a series of points approaching the solution from inside the feasible region are found by solving a sequence of subproblems characterized by an increasing positive parameter. We use preconditioned conjugate gradient (PCG) algorithm to solve the subproblems directly. The nonnegativity constraint is enforced by bend line search. The exact expression of the TV functional is used in our calculations. Simulation results show that the algorithm converges fast and the convergence is insensitive to the values of the regularization and reconstruction parameters.
Directory of Open Access Journals (Sweden)
Shkvarko Yuriy
2006-01-01
Full Text Available We address a new approach to solve the ill-posed nonlinear inverse problem of high-resolution numerical reconstruction of the spatial spectrum pattern (SSP of the backscattered wavefield sources distributed over the remotely sensed scene. An array or synthesized array radar (SAR that employs digital data signal processing is considered. By exploiting the idea of combining the statistical minimum risk estimation paradigm with numerical descriptive regularization techniques, we address a new fused statistical descriptive regularization (SDR strategy for enhanced radar imaging. Pursuing such an approach, we establish a family of the SDR-related SSP estimators, that encompass a manifold of existing beamforming techniques ranging from traditional matched filter to robust and adaptive spatial filtering, and minimum variance methods.
Directory of Open Access Journals (Sweden)
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.
Numerical simulation of the regularized long wave equation by He's homotopy perturbation method
Energy Technology Data Exchange (ETDEWEB)
Inc, Mustafa [Department of Mathematics, Firat University, 23119 Elazig (Turkey)], E-mail: minc@firat.edu.tr; Ugurlu, Yavuz [Department of Mathematics, Firat University, 23119 Elazig (Turkey)
2007-09-17
In this Letter, we present the homotopy perturbation method (shortly HPM) for obtaining the numerical solution of the RLW equation. We obtain the exact and numerical solutions of the Regularized Long Wave (RLW) equation for certain initial condition. The initial approximation can be freely chosen with possible unknown constants which can be determined by imposing the boundary and initial conditions. Comparison of the results with those of other methods have led us to significant consequences. The numerical solutions are compared with the known analytical solutions.
Numerical simulation of the regularized long wave equation by He's homotopy perturbation method
International Nuclear Information System (INIS)
Inc, Mustafa; Ugurlu, Yavuz
2007-01-01
In this Letter, we present the homotopy perturbation method (shortly HPM) for obtaining the numerical solution of the RLW equation. We obtain the exact and numerical solutions of the Regularized Long Wave (RLW) equation for certain initial condition. The initial approximation can be freely chosen with possible unknown constants which can be determined by imposing the boundary and initial conditions. Comparison of the results with those of other methods have led us to significant consequences. The numerical solutions are compared with the known analytical solutions
kernel oil by lipolytic organisms
African Journals Online (AJOL)
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 regularized vortex-particle mesh method for large eddy simulation
Spietz, H. J.; Walther, J. H.; Hejlesen, M. M.
2017-11-01
We present recent developments of the remeshed vortex particle-mesh method for simulating incompressible fluid flow. The presented method relies on a parallel higher-order FFT based solver for the Poisson equation. Arbitrary high order is achieved through regularization of singular Green's function solutions to the Poisson equation and recently we have derived novel high order solutions for a mixture of open and periodic domains. With this approach the simulated variables may formally be viewed as the approximate solution to the filtered Navier Stokes equations, hence we use the method for Large Eddy Simulation by including a dynamic subfilter-scale model based on test-filters compatible with the aforementioned regularization functions. Further the subfilter-scale model uses Lagrangian averaging, which is a natural candidate in light of the Lagrangian nature of vortex particle methods. A multiresolution variation of the method is applied to simulate the benchmark problem of the flow past a square cylinder at Re = 22000 and the obtained results are compared to results from the literature.
Optimized star sensors laboratory calibration method using a regularization neural network.
Zhang, Chengfen; Niu, Yanxiong; Zhang, Hao; Lu, Jiazhen
2018-02-10
High-precision ground calibration is essential to ensure the performance of star sensors. However, the complex distortion and multi-error coupling have brought great difficulties to traditional calibration methods, especially for large field of view (FOV) star sensors. Although increasing the complexity of models is an effective way to improve the calibration accuracy, it significantly increases the demand for calibration data. In order to achieve high-precision calibration of star sensors with large FOV, a novel laboratory calibration method based on a regularization neural network is proposed. A multi-layer structure neural network is designed to represent the mapping of the star vector and the corresponding star point coordinate directly. To ensure the generalization performance of the network, regularization strategies are incorporated into the net structure and the training algorithm. Simulation and experiment results demonstrate that the proposed method can achieve high precision with less calibration data and without any other priori information. Compared with traditional methods, the calibration error of the star sensor decreased by about 30%. The proposed method can satisfy the precision requirement for large FOV star sensors.
Credit scoring analysis using kernel discriminant
Widiharih, T.; Mukid, M. A.; Mustafid
2018-05-01
Credit scoring model is an important tool for reducing the risk of wrong decisions when granting credit facilities to applicants. This paper investigate the performance of kernel discriminant model in assessing customer credit risk. Kernel discriminant analysis is a non- parametric method which means that it does not require any assumptions about the probability distribution of the input. The main ingredient is a kernel that allows an efficient computation of Fisher discriminant. We use several kernel such as normal, epanechnikov, biweight, and triweight. The models accuracy was compared each other using data from a financial institution in Indonesia. The results show that kernel discriminant can be an alternative method that can be used to determine who is eligible for a credit loan. In the data we use, it shows that a normal kernel is relevant to be selected for credit scoring using kernel discriminant model. Sensitivity and specificity reach to 0.5556 and 0.5488 respectively.
Kernel parameter dependence in spatial factor analysis
DEFF Research Database (Denmark)
Nielsen, Allan Aasbjerg
2010-01-01
kernel PCA. Shawe-Taylor and Cristianini [4] is an excellent reference for kernel methods in general. Bishop [5] and Press et al. [6] describe kernel methods among many other subjects. 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 a kernel version of maximum autocorrelation factor (MAF) [7, 8] analysis to irregularly sampled stream sediment geochemistry data from South Greenland and illustrate the dependence...... of the kernel width. The 2,097 samples each covering on average 5 km2 are analyzed chemically for the content of 41 elements....
Differential regularization and renormalization: a new method of calculation in quantum field theory
International Nuclear Information System (INIS)
Freedman, D.Z.; Johnson, K.; Latorre, J.I.
1992-01-01
Most primitively divergent Feynman diagrams are well defined in x-space but too singular at short distances for transformation to p-space. A new method of regularization is developed in which singular functions are written as derivatives of less singular functions which contain a logarithmic mass scale. The Fourier transform is then defined by formal integration by parts. The procedure is extended to graphs with divergent subgraphs. No explicit cutoff or counterterms are required, and the method automatically delivers renormalized amplitudes which satisfy Callan-Symanzik equations. These features are thoroughly explored in massless φ 4 theory through 3-loop order, and the method yields explicit functional forms for all amplitudes with less difficulty than conventional methods which use dimensional regularization in p-space. The procedure also appears to be compatible with gauge invariance and the chiral structure of the standard model. This aspect is tested in extensive 1-loop calculations which include the Ward identity in quantum electrodynamics, the chiral anomaly, and the background field algorithm in non-abelian gauge theories. (orig.)
Directory of Open Access Journals (Sweden)
Shane Stimpson
2017-09-01
Full Text Available An essential component of the neutron transport solver is the resonance self-shielding calculation used to determine equivalence cross sections. The neutron transport code, MPACT, is currently using the subgroup self-shielding method, in which the method of characteristics (MOC is used to solve purely absorbing fixed-source problems. Recent efforts incorporating multigroup kernels to the MOC solvers in MPACT have reduced runtime by roughly 2×. Applying the same concepts for self-shielding and developing a novel lumped parameter approach to MOC, substantial improvements have also been made to the self-shielding computational efficiency without sacrificing any accuracy. These new multigroup and lumped parameter capabilities have been demonstrated on two test cases: (1 a single lattice with quarter symmetry known as VERA (Virtual Environment for Reactor Applications Progression Problem 2a and (2 a two-dimensional quarter-core slice known as Problem 5a-2D. From these cases, self-shielding computational time was reduced by roughly 3–4×, with a corresponding 15–20% increase in overall memory burden. An azimuthal angle sensitivity study also shows that only half as many angles are needed, yielding an additional speedup of 2×. In total, the improvements yield roughly a 7–8× speedup. Given these performance benefits, these approaches have been adopted as the default in MPACT.
International Nuclear Information System (INIS)
Stimpson, Shane G.; Liu, Yuxuan; Collins, Benjamin S.; Clarno, Kevin T.
2017-01-01
An essential component of the neutron transport solver is the resonance self-shielding calculation used to determine equivalence cross sections. The neutron transport code, MPACT, is currently using the subgroup self-shielding method, in which the method of characteristics (MOC) is used to solve purely absorbing fixed-source problems. Recent efforts incorporating multigroup kernels to the MOC solvers in MPACT have reduced runtime by roughly 2×. Applying the same concepts for self-shielding and developing a novel lumped parameter approach to MOC, substantial improvements have also been made to the self-shielding computational efficiency without sacrificing any accuracy. These new multigroup and lumped parameter capabilities have been demonstrated on two test cases: (1) a single lattice with quarter symmetry known as VERA (Virtual Environment for Reactor Applications) Progression Problem 2a and (2) a two-dimensional quarter-core slice known as Problem 5a-2D. From these cases, self-shielding computational time was reduced by roughly 3–4×, with a corresponding 15–20% increase in overall memory burden. An azimuthal angle sensitivity study also shows that only half as many angles are needed, yielding an additional speedup of 2×. In total, the improvements yield roughly a 7–8× speedup. Furthermore given these performance benefits, these approaches have been adopted as the default in MPACT.
Backtracking-Based Iterative Regularization Method for Image Compressive Sensing Recovery
Directory of Open Access Journals (Sweden)
Lingjun Liu
2017-01-01
Full Text Available This paper presents a variant of the iterative shrinkage-thresholding (IST algorithm, called backtracking-based adaptive IST (BAIST, for image compressive sensing (CS reconstruction. For increasing iterations, IST usually yields a smoothing of the solution and runs into prematurity. To add back more details, the BAIST method backtracks to the previous noisy image using L2 norm minimization, i.e., minimizing the Euclidean distance between the current solution and the previous ones. Through this modification, the BAIST method achieves superior performance while maintaining the low complexity of IST-type methods. Also, BAIST takes a nonlocal regularization with an adaptive regularizor to automatically detect the sparsity level of an image. Experimental results show that our algorithm outperforms the original IST method and several excellent CS techniques.
Tikhonov regularization method for the numerical inversion of Mellin transforms using splines
International Nuclear Information System (INIS)
Iqbal, M.
2005-01-01
Mellin transform is an ill-posed problem. These problems arise in many branches of science and engineering. In the typical situation one is interested in recovering the original function, given a finite number of noisy measurements of data. In this paper, we shall convert Mellin transform to Laplace transform and then an integral equation of the first kind of convolution type. We solve the integral equation using Tikhonov regularization with splines as basis function. The method is applied to various test examples in the literature and results are shown in the table
Regularization of the Fourier series of discontinuous functions by various summation methods
Energy Technology Data Exchange (ETDEWEB)
Ahmad, S.S.; Beghi, L. (Padua Univ. (Italy). Seminario Matematico)
1983-07-01
In this paper the regularization by various summation methods of the Fourier series of functions containing discontinuities of the first and second kind are studied and the results of the numerical analyses referring to some typical periodic functions are presented. In addition to the Cesaro and Lanczos weightings, a new (i.e. cosine) weighting for accelerating the convergence rate is proposed. A comparison with the results obtained by Garibotti and Massaro with the punctual Pade approximants (PPA) technique in case of a periodic step function is also carried out.
DEFF Research Database (Denmark)
Nielsen, Allan Aasbjerg
2007-01-01
This paper describes new extensions to the previously published multivariate alteration detection (MAD) method for change detection in bi-temporal, multi- and hypervariate data such as remote sensing imagery. Much like boosting methods often applied in data mining work, the iteratively reweighted...... to observations that show little change, i.e., for which the sum of squared, standardized MAD variates is small, and small weights are assigned to observations for which the sum is large. Like the original MAD method, the iterative extension is invariant to linear (affine) transformations of the original...... an agricultural region in Kenya, and hyperspectral airborne HyMap data from a small rural area in southeastern Germany are given. The latter case demonstrates the need for regularization....
International Nuclear Information System (INIS)
Ma, Zhongqing; Chen, Dengyu; Gu, Jie; Bao, Binfu; Zhang, Qisheng
2015-01-01
Highlights: • Model-free integral kinetics method and analytical TGA–FTIR were conducted on pyrolysis process of PKS. • The pyrolysis mechanism of PKS was elaborated. • Thermal stability was established: lignin > cellulose > xylan. • Detailed compositions in the volatiles of PKS pyrolysis were determinated. • The interaction of biomass three components led to the fluctuation of activation energy in PKS pyrolysis. - Abstract: Palm kernel shell (PKS) from palm oil production is a potential biomass source for bio-energy production. A fundamental understanding of PKS pyrolysis behavior and kinetics is essential to its efficient thermochemical conversion. The thermal degradation profile in derivative thermogravimetry (DTG) analysis shown two significant mass-loss peaks mainly related to the decomposition of hemicellulose and cellulose respectively. This characteristic differentiated with other biomass (e.g. wheat straw and corn stover) presented just one peak or accompanied with an extra “shoulder” peak (e.g. wheat straw). According to the Fourier transform infrared spectrometry (FTIR) analysis, the prominent volatile components generated by the pyrolysis of PKS were CO 2 (2400–2250 cm −1 and 586–726 cm −1 ), aldehydes, ketones, organic acids (1900–1650 cm −1 ), and alkanes, phenols (1475–1000 cm −1 ). The activation energy dependent on the conversion rate was estimated by two model-free integral methods: Flynn–Wall–Ozawa (FWO) and Kissinger–Akahira–Sunose (KAS) method at different heating rates. The fluctuation of activation energy can be interpreted as a result of interactive reactions related to cellulose, hemicellulose and lignin degradation, occurred in the pyrolysis process. Based on TGA–FTIR analysis and model free integral kinetics method, the pyrolysis mechanism of PKS was elaborated in this paper
The analytic regularization ζ function method and the cut-off method in Casimir effect
International Nuclear Information System (INIS)
Svaiter, N.F.; Svaiter, B.F.
1990-01-01
The zero point energy associated to a hermitian massless scalar field in the presence of perfectly reflecting plates in a three dimensional flat space-time is discussed. A new technique to unify two different methods - the ζ function and a variant of the cut-off method - used to obtain the so called Casimir energy is presented, and the proof of the analytic equivalence between both methods is given. (author)
Classification of Hyperspectral Images Using Kernel Fully Constrained Least Squares
Directory of Open Access Journals (Sweden)
Jianjun Liu
2017-11-01
Full Text Available As a widely used classifier, sparse representation classification (SRC has shown its good performance for hyperspectral image classification. Recent works have highlighted that it is the collaborative representation mechanism under SRC that makes SRC a highly effective technique for classification purposes. If the dimensionality and the discrimination capacity of a test pixel is high, other norms (e.g., ℓ 2 -norm can be used to regularize the coding coefficients, except for the sparsity ℓ 1 -norm. In this paper, we show that in the kernel space the nonnegative constraint can also play the same role, and thus suggest the investigation of kernel fully constrained least squares (KFCLS for hyperspectral image classification. Furthermore, in order to improve the classification performance of KFCLS by incorporating spatial-spectral information, we investigate two kinds of spatial-spectral methods using two regularization strategies: (1 the coefficient-level regularization strategy, and (2 the class-level regularization strategy. Experimental results conducted on four real hyperspectral images demonstrate the effectiveness of the proposed KFCLS, and show which way to incorporate spatial-spectral information efficiently in the regularization framework.
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.
International Nuclear Information System (INIS)
Soussaline, F.; LeCoq, C.; Raynaud, C.; Kellershohn
1982-01-01
The potential of the Regularizing Iterative Method (RIM), when used in brain studies, is evaluated. RIM is designed to provide fast and accurate reconstruction of tomographic images when non-uniform attenuation is to be accounted for. As indicated by phantom studies, this method improves the contrast and the signal-to-noise ratio as compared to those obtained with Filtered Back Projection (FBP) technique. Preliminary results obtained in brain studies using isopropil-amphetamine I-123 (AMPI-123) are very encouraging in terms of quantitative regional cellular activity. However, the clinical usefulness of this mathematically accurate reconstruction procedure is going to be demonstrated, in comparing quantitative data in heart or liver studies where control values can be obtained
Sitompul, Monica Angelina
2015-01-01
Have been conducted Determination of Iodin Value by method titration to some Hydrogenated Palm Kernel Oil (HPKO) and Refined Bleached Deodorized Palm Kernel Oil (RBDPKO). The result of analysis obtained the Iodin Value in Hydrogenated Palm Kernel Oil (A) = 0,16 gr I2/100gr, Hydrogenated Palm Kernel Oil (B) = 0,20 gr I2/100gr, Hydrogenated Palm Kernel Oil (C) = 0,24 gr I2/100gr. And in Refined Bleached Deodorized Palm Kernel Oil (A) = 17,51 gr I2/100gr, Refined Bleached Deodorized Palm Kernel ...
Directory of Open Access Journals (Sweden)
H. O. Bakodah
2013-01-01
Full Text Available A method of lines approach to the numerical solution of nonlinear wave equations typified by the regularized long wave (RLW is presented. The method developed uses a finite differences discretization to the space. Solution of the resulting system was obtained by applying fourth Runge-Kutta time discretization method. Using Von Neumann stability analysis, it is shown that the proposed method is marginally stable. To test the accuracy of the method some numerical experiments on test problems are presented. Test problems including solitary wave motion, two-solitary wave interaction, and the temporal evaluation of a Maxwellian initial pulse are studied. The accuracy of the present method is tested with and error norms and the conservation properties of mass, energy, and momentum under the RLW equation.
AN AUTOMATED METHOD FOR 3D ROOF OUTLINE GENERATION AND REGULARIZATION IN AIRBONE LASER SCANNER DATA
Directory of Open Access Journals (Sweden)
S. N. Perera
2012-07-01
Full Text Available In this paper, an automatic approach for the generation and regularization of 3D roof boundaries in Airborne Laser scanner data is presented. The workflow is commenced by segmentation of the point clouds. A classification step and a rule based roof extraction step are followed the planar segmentation. Refinement on roof extraction is performed in order to minimize the effect due to urban vegetation. Boundary points of the connected roof planes are extracted and fitted series of straight line segments. Each line is then regularized with respect to the dominant building orientation. We introduce the usage of cycle graphs for the best use of topological information. Ridge-lines and step-edges are basically extracted to recognise correct topological relationships among the roof faces. Inner roof corners are geometrically fitted based on the closed cycle graphs. Outer boundary is reconstructed using the same concept but with the outer most cycle graph. In here, union of the sub cycles is taken. Intermediate line segments (outer bounds are intersected to reconstruct the roof eave lines. Two test areas with two different point densities are tested with the developed approach. Performance analysis of the test results is provided to demonstrate the applicability of the method.
Hermite regularization of the lattice Boltzmann method for open source computational aeroacoustics.
Brogi, F; Malaspinas, O; Chopard, B; Bonadonna, C
2017-10-01
The lattice Boltzmann method (LBM) is emerging as a powerful engineering tool for aeroacoustic computations. However, the LBM has been shown to present accuracy and stability issues in the medium-low Mach number range, which is of interest for aeroacoustic applications. Several solutions have been proposed but are often too computationally expensive, do not retain the simplicity and the advantages typical of the LBM, or are not described well enough to be usable by the community due to proprietary software policies. An original regularized collision operator is proposed, based on the expansion of Hermite polynomials, that greatly improves the accuracy and stability of the LBM without significantly altering its algorithm. The regularized LBM can be easily coupled with both non-reflective boundary conditions and a multi-level grid strategy, essential ingredients for aeroacoustic simulations. Excellent agreement was found between this approach and both experimental and numerical data on two different benchmarks: the laminar, unsteady flow past a 2D cylinder and the 3D turbulent jet. Finally, most of the aeroacoustic computations with LBM have been done with commercial software, while here the entire theoretical framework is implemented using an open source library (palabos).
Regular pipeline maintenance of gas pipeline using technical operational diagnostics methods
Energy Technology Data Exchange (ETDEWEB)
Volentic, J [Gas Transportation Department, Slovensky plynarensky priemysel, Slovak Gas Industry, Bratislava (Slovakia)
1998-12-31
Slovensky plynarensky priemysel (SPP) has operated 17 487 km of gas pipelines in 1995. The length of the long-line pipelines reached 5 191 km, distribution network was 12 296 km. The international transit system of long-line gas pipelines ranged 1 939 km of pipelines of various dimensions. The described scale of transport and distribution system represents a multibillion investments stored in the ground, which are exposed to the environmental influences and to pipeline operational stresses. In spite of all technical and maintenance arrangements, which have to be performed upon operating gas pipelines, the gradual ageing takes place anyway, expressed in degradation process both in steel tube, as well as in the anti-corrosion coating. Within a certain time horizon, a consistent and regular application of methods and means of in-service technical diagnostics and rehabilitation of existing pipeline systems make it possible to save substantial investment funds, postponing the need in funds for a complex or partial reconstruction or a new construction of a specific gas section. The purpose of this presentation is to report on the implementation of the programme of in-service technical diagnostics of gas pipelines within the framework of regular maintenance of SPP s.p. Bratislava high pressure gas pipelines. (orig.) 6 refs.
Regular pipeline maintenance of gas pipeline using technical operational diagnostics methods
Energy Technology Data Exchange (ETDEWEB)
Volentic, J. [Gas Transportation Department, Slovensky plynarensky priemysel, Slovak Gas Industry, Bratislava (Slovakia)
1997-12-31
Slovensky plynarensky priemysel (SPP) has operated 17 487 km of gas pipelines in 1995. The length of the long-line pipelines reached 5 191 km, distribution network was 12 296 km. The international transit system of long-line gas pipelines ranged 1 939 km of pipelines of various dimensions. The described scale of transport and distribution system represents a multibillion investments stored in the ground, which are exposed to the environmental influences and to pipeline operational stresses. In spite of all technical and maintenance arrangements, which have to be performed upon operating gas pipelines, the gradual ageing takes place anyway, expressed in degradation process both in steel tube, as well as in the anti-corrosion coating. Within a certain time horizon, a consistent and regular application of methods and means of in-service technical diagnostics and rehabilitation of existing pipeline systems make it possible to save substantial investment funds, postponing the need in funds for a complex or partial reconstruction or a new construction of a specific gas section. The purpose of this presentation is to report on the implementation of the programme of in-service technical diagnostics of gas pipelines within the framework of regular maintenance of SPP s.p. Bratislava high pressure gas pipelines. (orig.) 6 refs.
Provencher, Stephen W.
1982-09-01
CONTIN is a portable Fortran IV package for inverting noisy linear operator equations. These problems occur in the analysis of data from a wide variety experiments. They are generally ill-posed problems, which means that errors in an unregularized inversion are unbounded. Instead, CONTIN seeks the optimal solution by incorporating parsimony and any statistical prior knowledge into the regularizor and absolute prior knowledge into equallity and inequality constraints. This can be greatly increase the resolution and accuracyh of the solution. CONTIN is very flexible, consisting of a core of about 50 subprograms plus 13 small "USER" subprograms, which the user can easily modify to specify special-purpose constraints, regularizors, operator equations, simulations, statistical weighting, etc. Specjial collections of USER subprograms are available for photon correlation spectroscopy, multicomponent spectra, and Fourier-Bessel, Fourier and Laplace transforms. Numerically stable algorithms are used throughout CONTIN. A fairly precise definition of information content in terms of degrees of freedom is given. The regularization parameter can be automatically chosen on the basis of an F-test and confidence region. The interpretation of the latter and of error estimates based on the covariance matrix of the constrained regularized solution are discussed. The strategies, methods and options in CONTIN are outlined. The program itself is described in the following paper.
Generalized Bregman distances and convergence rates for non-convex regularization methods
International Nuclear Information System (INIS)
Grasmair, Markus
2010-01-01
We generalize the notion of Bregman distance using concepts from abstract convexity in order to derive convergence rates for Tikhonov regularization with non-convex regularization terms. In particular, we study the non-convex regularization of linear operator equations on Hilbert spaces, showing that the conditions required for the application of the convergence rates results are strongly related to the standard range conditions from the convex case. Moreover, we consider the setting of sparse regularization, where we show that a rate of order δ 1/p holds, if the regularization term has a slightly faster growth at zero than |t| p
Bukhari, Hassan J.
2017-12-01
In this paper a framework for robust optimization of mechanical design problems and process systems that have parametric uncertainty is presented using three different approaches. Robust optimization problems are formulated so that the optimal solution is robust which means it is minimally sensitive to any perturbations in parameters. The first method uses the price of robustness approach which assumes the uncertain parameters to be symmetric and bounded. The robustness for the design can be controlled by limiting the parameters that can perturb.The second method uses the robust least squares method to determine the optimal parameters when data itself is subjected to perturbations instead of the parameters. The last method manages uncertainty by restricting the perturbation on parameters to improve sensitivity similar to Tikhonov regularization. The methods are implemented on two sets of problems; one linear and the other non-linear. This methodology will be compared with a prior method using multiple Monte Carlo simulation runs which shows that the approach being presented in this paper results in better performance.
Energy Technology Data Exchange (ETDEWEB)
Bobodzhanov, A A; Safonov, V F [National Research University " Moscow Power Engineering Institute" , Moscow (Russian Federation)
2013-07-31
The paper deals with extending the Lomov regularization method to classes of singularly perturbed Fredholm-type integro-differential systems, which have not so far been studied. In these the limiting operator is discretely noninvertible. Such systems are commonly known as problems with unstable spectrum. Separating out the essential singularities in the solutions to these problems presents great difficulties. The principal one is to give an adequate description of the singularities induced by 'instability points' of the spectrum. A methodology for separating singularities by using normal forms is developed. It is applied to the above type of systems and is substantiated in these systems. Bibliography: 10 titles.
Nararidh, Niti
2013-11-01
Choanoflagellates are unicellular organisms whose intriguing morphology includes a set of collars/microvilli emanating from the cell body, surrounding the beating flagellum. We investigated the role of the microvilli in the feeding and swimming behavior of the organism using a three-dimensional model based on the method of regularized Stokeslets. This model allows us to examine the velocity generated around the feeding organism tethered in place, as well as to predict the paths of surrounding free flowing particles. In particular, we can depict the effective capture of nutritional particles and bacteria in the fluid, showing the hydrodynamic cooperation between the cell, flagellum, and microvilli of the organism. Funding Source: Murchison Undergraduate Research Fellowship.
REGULAR METHOD FOR SYNTHESIS OF BASIC BENT-SQUARES OF RANDOM ORDER
Directory of Open Access Journals (Sweden)
A. V. Sokolov
2016-01-01
Full Text Available The paper is devoted to the class construction of the most non-linear Boolean bent-functions of any length N = 2k (k = 2, 4, 6…, on the basis of their spectral representation – Agievich bent squares. These perfect algebraic constructions are used as a basis to build many new cryptographic primitives, such as generators of pseudo-random key sequences, crypto graphic S-boxes, etc. Bent-functions also find their application in the construction of C-codes in the systems with code division multiple access (CDMA to provide the lowest possible value of Peak-to-Average Power Ratio (PAPR k = 1, as well as for the construction of error-correcting codes and systems of orthogonal biphasic signals. All the numerous applications of bent-functions relate to the theory of their synthesis. However, regular methods for complete class synthesis of bent-functions of any length N = 2k are currently unknown. The paper proposes a regular synthesis method for the basic Agievich bent squares of any order n, based on a regular operator of dyadic shift. Classification for a complete set of spectral vectors of lengths (l = 8, 16, … based on a criterion of the maximum absolute value and set of absolute values of spectral components has been carried out in the paper. It has been shown that any spectral vector can be a basis for building bent squares. Results of the synthesis for the Agievich bent squares of order n = 8 have been generalized and it has been revealed that there are only 3 basic bent squares for this order, while the other 5 can be obtained with help the operation of step-cyclic shift. All the basic bent squares of order n = 16 have been synthesized that allows to construct the bent-functions of length N = 256. The obtained basic bent squares can be used either for direct synthesis of bent-functions and their practical application or for further research in order to synthesize new structures of bent squares of orders n = 16, 32, 64, …
Yong, Peng; Liao, Wenyuan; Huang, Jianping; Li, Zhenchuan
2018-04-01
Full waveform inversion is an effective tool for recovering the properties of the Earth from seismograms. However, it suffers from local minima caused mainly by the limited accuracy of the starting model and the lack of a low-frequency component in the seismic data. Because of the high velocity contrast between salt and sediment, the relation between the waveform and velocity perturbation is strongly nonlinear. Therefore, salt inversion can easily get trapped in the local minima. Since the velocity of salt is nearly constant, we can make the most of this characteristic with total variation regularization to mitigate the local minima. In this paper, we develop an adaptive primal dual hybrid gradient method to implement total variation regularization by projecting the solution onto a total variation norm constrained convex set, through which the total variation norm constraint is satisfied at every model iteration. The smooth background velocities are first inverted and the perturbations are gradually obtained by successively relaxing the total variation norm constraints. Numerical experiment of the projection of the BP model onto the intersection of the total variation norm and box constraints has demonstrated the accuracy and efficiency of our adaptive primal dual hybrid gradient method. A workflow is designed to recover complex salt structures in the BP 2004 model and the 2D SEG/EAGE salt model, starting from a linear gradient model without using low-frequency data below 3 Hz. The salt inversion processes demonstrate that wavefield reconstruction inversion with a total variation norm and box constraints is able to overcome local minima and inverts the complex salt velocity layer by layer.
Chen, Haiwen; Holland, Paul
2010-01-01
In this paper, we develop a new curvilinear equating for the nonequivalent groups with anchor test (NEAT) design under the assumption of the classical test theory model, that we name curvilinear Levine observed score equating. In fact, by applying both the kernel equating framework and the mean preserving linear transformation of…
Zhu, Xiaofeng; Suk, Heung-Il; Wang, Li; Lee, Seong-Whan; Shen, Dinggang
2017-05-01
In this paper, we focus on joint regression and classification for Alzheimer's disease diagnosis and propose a new feature selection method by embedding the relational information inherent in the observations into a sparse multi-task learning framework. Specifically, the relational information includes three kinds of relationships (such as feature-feature relation, response-response relation, and sample-sample relation), for preserving three kinds of the similarity, such as for the features, the response variables, and the samples, respectively. To conduct feature selection, we first formulate the objective function by imposing these three relational characteristics along with an ℓ 2,1 -norm regularization term, and further propose a computationally efficient algorithm to optimize the proposed objective function. With the dimension-reduced data, we train two support vector regression models to predict the clinical scores of ADAS-Cog and MMSE, respectively, and also a support vector classification model to determine the clinical label. We conducted extensive experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to validate the effectiveness of the proposed method. Our experimental results showed the efficacy of the proposed method in enhancing the performances of both clinical scores prediction and disease status identification, compared to the state-of-the-art methods. Copyright © 2015 Elsevier B.V. All rights reserved.
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.
Response matrix of regular moderator volumes with 3He detector using Monte Carlo methods
International Nuclear Information System (INIS)
Baltazar R, A.; Vega C, H. R.; Ortiz R, J. M.; Solis S, L. O.; Castaneda M, R.; Soto B, T. G.; Medina C, D.
2017-10-01
In the last three decades the uses of Monte Carlo methods, for the estimation of physical phenomena associated with the interaction of radiation with matter, have increased considerably. The reason is due to the increase in computing capabilities and the reduction of computer prices. Monte Carlo methods allow modeling and simulating real systems before their construction, saving time and costs. The interaction mechanisms between neutrons and matter are diverse and range from elastic dispersion to nuclear fission; to facilitate the neutrons detection, is necessary to moderate them until reaching electronic equilibrium with the medium at standard conditions of pressure and temperature, in this state the total cross section of the 3 He is large. The objective of the present work was to estimate the response matrix of a proportional detector of 3 He using regular volumes of moderator through Monte Carlo methods. Neutron monoenergetic sources with energies of 10 -9 to 20 MeV and polyethylene moderators of different sizes were used. The calculations were made with the MCNP5 code; the number of stories for each detector-moderator combination was large enough to obtain errors less than 1.5%. We found that for small moderators the highest response is obtained for lower energy neutrons, when increasing the moderator dimension we observe that the response decreases for neutrons of lower energy and increases for higher energy neutrons. The total sum of the responses of each moderator allows obtaining a response close to a constant function. (Author)
Clustering via Kernel Decomposition
DEFF Research Database (Denmark)
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....
Gaussian interaction profile kernels for predicting drug-target interaction.
van Laarhoven, Twan; Nabuurs, Sander B; Marchiori, Elena
2011-11-01
The in silico prediction of potential interactions between drugs and target proteins is of core importance for the identification of new drugs or novel targets for existing drugs. However, only a tiny portion of all drug-target pairs in current datasets are experimentally validated interactions. This motivates the need for developing computational methods that predict true interaction pairs with high accuracy. We show that a simple machine learning method that uses the drug-target network as the only source of information is capable of predicting true interaction pairs with high accuracy. Specifically, we introduce interaction profiles of drugs (and of targets) in a network, which are binary vectors specifying the presence or absence of interaction with every target (drug) in that network. We define a kernel on these profiles, called the Gaussian Interaction Profile (GIP) kernel, and use a simple classifier, (kernel) Regularized Least Squares (RLS), for prediction drug-target interactions. We test comparatively the effectiveness of RLS with the GIP kernel on four drug-target interaction networks used in previous studies. The proposed algorithm achieves area under the precision-recall curve (AUPR) up to 92.7, significantly improving over results of state-of-the-art methods. Moreover, we show that using also kernels based on chemical and genomic information further increases accuracy, with a neat improvement on small datasets. These results substantiate the relevance of the network topology (in the form of interaction profiles) as source of information for predicting drug-target interactions. Software and Supplementary Material are available at http://cs.ru.nl/~tvanlaarhoven/drugtarget2011/. tvanlaarhoven@cs.ru.nl; elenam@cs.ru.nl. Supplementary data are available at Bioinformatics online.
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…
International Nuclear Information System (INIS)
Soussaline, F.; Bidaut, L.; Raynaud, C.; Le Coq, G.
1983-06-01
An analytical solution to the SPECT reconstruction problem, where the actual attenuation effect can be included, was developped using a regularizing iterative method (RIM). The potential of this approach in quantitative brain studies when using a tracer for cerebrovascular disorders is now under evaluation. Mathematical simulations for a distributed activity in the brain surrounded by the skull and physical phantom studies were performed, using a rotating camera based SPECT system, allowing the calibration of the system and the evaluation of the adapted method to be used. On the simulation studies, the contrast obtained along a profile, was less than 5%, the standard deviation 8% and the quantitative accuracy 13%, for a uniform emission distribution of mean = 100 per pixel and a double attenuation coefficient of μ = 0.115 cm -1 and 0.5 cm -1 . Clinical data obtained after injection of 123 I (AMPI) were reconstructed using the RIM without and with cerebrovascular diseases or lesion defects. Contour finding techniques were used for the delineation of the brain and the skull, and measured attenuation coefficients were assumed within these two regions. Using volumes of interest, selected on homogeneous regions on an hemisphere and reported symetrically, the statistical uncertainty for 300 K events in the tomogram was found to be 12%, the index of symetry was of 4% for normal distribution. These results suggest that quantitative SPECT reconstruction for brain distribution is feasible, and that combined with an adapted tracer and an adequate model physiopathological parameters could be extracted
International Nuclear Information System (INIS)
Soussaline, Francoise; Cao, A.; Lecoq, G.
1981-06-01
An analytically exact solution to the attenuated tomographic operator is proposed. Such a technique called Regularizing Iterative Method (RIM) belongs to the iterative class of procedures where a priori knowledge can be introduced on the evaluation of the size and shape of the activity domain to be reconstructed, and on the exact attenuation distribution. The relaxation factor used is so named because it leads to fast convergence and provides noise filtering for a small number of iteractions. The effectiveness of such a method was tested in the Single Photon Emission Computed Tomography (SPECT) reconstruction problem, with the goal of precise correction for attenuation before quantitative study. Its implementation involves the use of a rotating scintillation camera based SPECT detector connected to a mini computer system. Mathematical simulations of cylindical uniformly attenuated phantoms indicate that in the range of a priori calculated relaxation factor a fast converging solution can always be found with a (contrast) accuracy of the order of 0.2 to 4% given that numerical errors and noise are or not, taken into account. The sensitivity of the (RIM) algorithm to errors in the size of the reconstructed object and in the value of the attenuation coefficient μ was studied, using the same simulation data. Extreme variations of +- 15% in these parameters will lead to errors of the order of +- 20% in the quantitative results. Physical phantoms representing a variety of geometrical situations were also studied
A multi-scale kernel bundle for LDDMM
DEFF Research Database (Denmark)
Sommer, Stefan Horst; Nielsen, Mads; Lauze, Francois Bernard
2011-01-01
The Large Deformation Diffeomorphic Metric Mapping framework constitutes a widely used and mathematically well-founded setup for registration in medical imaging. At its heart lies the notion of the regularization kernel, and the choice of kernel greatly affects the results of registrations...
Regularization and computational methods for precise solution of perturbed orbit transfer problems
Woollands, Robyn Michele
The author has developed a suite of algorithms for solving the perturbed Lambert's problem in celestial mechanics. These algorithms have been implemented as a parallel computation tool that has broad applicability. This tool is composed of four component algorithms and each provides unique benefits for solving a particular type of orbit transfer problem. The first one utilizes a Keplerian solver (a-iteration) for solving the unperturbed Lambert's problem. This algorithm not only provides a "warm start" for solving the perturbed problem but is also used to identify which of several perturbed solvers is best suited for the job. The second algorithm solves the perturbed Lambert's problem using a variant of the modified Chebyshev-Picard iteration initial value solver that solves two-point boundary value problems. This method converges over about one third of an orbit and does not require a Newton-type shooting method and thus no state transition matrix needs to be computed. The third algorithm makes use of regularization of the differential equations through the Kustaanheimo-Stiefel transformation and extends the domain of convergence over which the modified Chebyshev-Picard iteration two-point boundary value solver will converge, from about one third of an orbit to almost a full orbit. This algorithm also does not require a Newton-type shooting method. The fourth algorithm uses the method of particular solutions and the modified Chebyshev-Picard iteration initial value solver to solve the perturbed two-impulse Lambert problem over multiple revolutions. The method of particular solutions is a shooting method but differs from the Newton-type shooting methods in that it does not require integration of the state transition matrix. The mathematical developments that underlie these four algorithms are derived in the chapters of this dissertation. For each of the algorithms, some orbit transfer test cases are included to provide insight on accuracy and efficiency of these
Void Structures in Regularly Patterned ZnO Nanorods Grown with the Hydrothermal Method
Directory of Open Access Journals (Sweden)
Yu-Feng Yao
2014-01-01
Full Text Available The void structures and related optical properties after thermal annealing with ambient oxygen in regularly patterned ZnO nanrorod (NR arrays grown with the hydrothermal method are studied. In increasing the thermal annealing temperature, void distribution starts from the bottom and extends to the top of an NR in the vertical (c-axis growth region. When the annealing temperature is higher than 400°C, void distribution spreads into the lateral (m-axis growth region. Photoluminescence measurement shows that the ZnO band-edge emission, in contrast to defect emission in the yellow-red range, is the strongest under the n-ZnO NR process conditions of 0.003 M in Ga-doping concentration and 300°C in thermal annealing temperature with ambient oxygen. Energy dispersive X-ray spectroscopy data indicate that the concentration of hydroxyl groups in the vertical growth region is significantly higher than that in the lateral growth region. During thermal annealing, hydroxyl groups are desorbed from the NR leaving anion vacancies for reacting with cation vacancies to form voids.
3D DC Resistivity Inversion with Topography Based on Regularized Conjugate Gradient Method
Directory of Open Access Journals (Sweden)
Jian-ke Qiang
2013-01-01
Full Text Available During the past decades, we observed a strong interest in 3D DC resistivity inversion and imaging with complex topography. In this paper, we implemented 3D DC resistivity inversion based on regularized conjugate gradient method with FEM. The Fréchet derivative is assembled with the electric potential in order to speed up the inversion process based on the reciprocity theorem. In this study, we also analyzed the sensitivity of the electric potential on the earth’s surface to the conductivity in each cell underground and introduced an optimized weighting function to produce new sensitivity matrix. The synthetic model study shows that this optimized weighting function is helpful to improve the resolution of deep anomaly. By incorporating topography into inversion, the artificial anomaly which is actually caused by topography can be eliminated. As a result, this algorithm potentially can be applied to process the DC resistivity data collected in mountain area. Our synthetic model study also shows that the convergence and computation speed are very stable and fast.
Ranking Support Vector Machine with Kernel Approximation
Directory of Open Access Journals (Sweden)
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.
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
Consistent Estimation of Pricing Kernels from Noisy Price Data
Vladislav Kargin
2003-01-01
If pricing kernels are assumed non-negative then the inverse problem of finding the pricing kernel is well-posed. The constrained least squares method provides a consistent estimate of the pricing kernel. When the data are limited, a new method is suggested: relaxed maximization of the relative entropy. This estimator is also consistent. Keywords: $\\epsilon$-entropy, non-parametric estimation, pricing kernel, inverse problems.
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 multiresolution method for solving the Poisson equation using high order regularization
DEFF Research Database (Denmark)
Hejlesen, Mads Mølholm; Walther, Jens Honore
2016-01-01
We present a novel high order multiresolution Poisson solver based on regularized Green's function solutions to obtain exact free-space boundary conditions while using fast Fourier transforms for computational efficiency. Multiresolution is a achieved through local refinement patches and regulari......We present a novel high order multiresolution Poisson solver based on regularized Green's function solutions to obtain exact free-space boundary conditions while using fast Fourier transforms for computational efficiency. Multiresolution is a achieved through local refinement patches...... and regularized Green's functions corresponding to the difference in the spatial resolution between the patches. The full solution is obtained utilizing the linearity of the Poisson equation enabling super-position of solutions. We show that the multiresolution Poisson solver produces convergence rates...
Hesford, Andrew J.; Waag, Robert C.
2010-10-01
The fast multipole method (FMM) is applied to the solution of large-scale, three-dimensional acoustic scattering problems involving inhomogeneous objects defined on a regular grid. The grid arrangement is especially well suited to applications in which the scattering geometry is not known a priori and is reconstructed on a regular grid using iterative inverse scattering algorithms or other imaging techniques. The regular structure of unknown scattering elements facilitates a dramatic reduction in the amount of storage and computation required for the FMM, both of which scale linearly with the number of scattering elements. In particular, the use of fast Fourier transforms to compute Green's function convolutions required for neighboring interactions lowers the often-significant cost of finest-level FMM computations and helps mitigate the dependence of FMM cost on finest-level box size. Numerical results demonstrate the efficiency of the composite method as the number of scattering elements in each finest-level box is increased.
Relationship between attenuation coefficients and dose-spread kernels
International Nuclear Information System (INIS)
Boyer, A.L.
1988-01-01
Dose-spread kernels can be used to calculate the dose distribution in a photon beam by convolving the kernel with the primary fluence distribution. The theoretical relationships between various types and components of dose-spread kernels relative to photon attenuation coefficients are explored. These relations can be valuable as checks on the conservation of energy by dose-spread kernels calculated by analytic or Monte Carlo methods
International Nuclear Information System (INIS)
Zhong Jian; Huang Si-Xun; Fei Jian-Fang; Du Hua-Dong; Zhang Liang
2011-01-01
According to the conclusion of the simulation experiments in paper I, the Tikhonov regularization method is applied to cyclone wind retrieval with a rain-effect-considering geophysical model function (called GMF+Rain). The GMF+Rain model which is based on the NASA scatterometer-2 (NSCAT2) GMF is presented to compensate for the effects of rain on cyclone wind retrieval. With the multiple solution scheme (MSS), the noise of wind retrieval is effectively suppressed, but the influence of the background increases. It will cause a large wind direction error in ambiguity removal when the background error is large. However, this can be mitigated by the new ambiguity removal method of Tikhonov regularization as proved in the simulation experiments. A case study on an extratropical cyclone of hurricane observed with SeaWinds at 25-km resolution shows that the retrieved wind speed for areas with rain is in better agreement with that derived from the best track analysis for the GMF+Rain model, but the wind direction obtained with the two-dimensional variational (2DVAR) ambiguity removal is incorrect. The new method of Tikhonov regularization effectively improves the performance of wind direction ambiguity removal through choosing appropriate regularization parameters and the retrieved wind speed is almost the same as that obtained from the 2DVAR. (electromagnetism, optics, acoustics, heat transfer, classical mechanics, and fluid dynamics)
DEFF Research Database (Denmark)
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...
Point-splitting as a regularization method for λφ4-type vertices: Abelian case
International Nuclear Information System (INIS)
Moura-Melo, Winder A.; Helayel Neto, J.A.
1998-11-01
We obtained regularized Abelian Lagrangians containing λφ 4 -type vertices by means of a suitable point-splitting procedure. The calculation is developed in details for a general Lagrangian, whose fields (gauge and matter ones) satisfy certain conditions. We illustrates our results by considering some special cases, such as the Abelian Higgs, the (ψ-barψ) 2 and the Avdeev-Chizov (real rank-2 antisymmetric tensor as matter fields) models. We also discuss some features of the obtained Lagrangian such as the regularity and non-locality of its new integrating terms. Moreover, the resolution of the Abelian case may teach us some useful technical aspects when dealing with the non-Abelian one. (author)
International Nuclear Information System (INIS)
Lee, H.C.; Milgram, M.S.
1984-07-01
A hybrid of dimensional and analytic regularization is used to regulate and uncover a Meijer's G-function representation for a class of massless, divergent Feynman integrals in an axial gauge. Integrals in the covariant gauge belong to a subclass and those in the light-cone gauge are reached by analytic continuation. The method decouples the physical ultraviolet and infrared singularities from the spurious axial gauge singularity but regulates all three simultaneously. For the axial gauge singularity, the new analytic method is more powerful and elegant than the old principal value prescription, but the two methods yield identical infinite as well as regular parts. It is shown that dimensional and analytic regularization can be made equivalent, implying that the former method is free from spurious γ5-anomalies and the latter preserves gauge invariance. The hybrid method permits the evaluation of integrals containing arbritrary integer powers of logarithms in the integrand by differentiation with respect to exponents. Such 'exponent derivatives' generate the same set of 'polylogs' as that generated in multi-loop integrals in perturbation theories and may be useful for solving equations in nonperturbation theories. The close relation between the method of exponent derivatives and the prescription of 't Hooft and Veltman for treating overlapping divergencies is pointed out. It is demonstrated that both methods generate functions that are free from unrecognizable logarithmic infinite parts. Nonperturbation theories expressed in terms of exponent derivatives are thus renormalizable. Some intriguing connections between nonperturbation theories and nonintegral exponents are pointed out
A Projection free method for Generalized Eigenvalue Problem with a nonsmooth Regularizer.
Hwang, Seong Jae; Collins, Maxwell D; Ravi, Sathya N; Ithapu, Vamsi K; Adluru, Nagesh; Johnson, Sterling C; Singh, Vikas
2015-12-01
Eigenvalue problems are ubiquitous in computer vision, covering a very broad spectrum of applications ranging from estimation problems in multi-view geometry to image segmentation. Few other linear algebra problems have a more mature set of numerical routines available and many computer vision libraries leverage such tools extensively. However, the ability to call the underlying solver only as a "black box" can often become restrictive. Many 'human in the loop' settings in vision frequently exploit supervision from an expert, to the extent that the user can be considered a subroutine in the overall system. In other cases, there is additional domain knowledge, side or even partial information that one may want to incorporate within the formulation. In general, regularizing a (generalized) eigenvalue problem with such side information remains difficult. Motivated by these needs, this paper presents an optimization scheme to solve generalized eigenvalue problems (GEP) involving a (nonsmooth) regularizer. We start from an alternative formulation of GEP where the feasibility set of the model involves the Stiefel manifold. The core of this paper presents an end to end stochastic optimization scheme for the resultant problem. We show how this general algorithm enables improved statistical analysis of brain imaging data where the regularizer is derived from other 'views' of the disease pathology, involving clinical measurements and other image-derived representations.
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%.
International Nuclear Information System (INIS)
Zhong Jian; Huang Si-Xun; Du Hua-Dong; Zhang Liang
2011-01-01
Scatterometer is an instrument which provides all-day and large-scale wind field information, and its application especially to wind retrieval always attracts meteorologists. Certain reasons cause large direction error, so it is important to find where the error mainly comes. Does it mainly result from the background field, the normalized radar cross-section (NRCS) or the method of wind retrieval? It is valuable to research. First, depending on SDP2.0, the simulated ‘true’ NRCS is calculated from the simulated ‘true’ wind through the geophysical model function NSCAT2. The simulated background field is configured by adding a noise to the simulated ‘true’ wind with the non-divergence constraint. Also, the simulated ‘measured’ NRCS is formed by adding a noise to the simulated ‘true’ NRCS. Then, the sensitivity experiments are taken, and the new method of regularization is used to improve the ambiguity removal with simulation experiments. The results show that the accuracy of wind retrieval is more sensitive to the noise in the background than in the measured NRCS; compared with the two-dimensional variational (2DVAR) ambiguity removal method, the accuracy of wind retrieval can be improved with the new method of Tikhonov regularization through choosing an appropriate regularization parameter, especially for the case of large error in the background. The work will provide important information and a new method for the wind retrieval with real data. (electromagnetism, optics, acoustics, heat transfer, classical mechanics, and fluid dynamics)
Energy Technology Data Exchange (ETDEWEB)
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`.
Prot, Olivier; SantolíK, OndřEj; Trotignon, Jean-Gabriel; Deferaudy, Hervé
2006-06-01
An entropy regularization algorithm (ERA) has been developed to compute the wave-energy density from electromagnetic field measurements. It is based on the wave distribution function (WDF) concept. To assess its suitability and efficiency, the algorithm is applied to experimental data that has already been analyzed using other inversion techniques. The FREJA satellite data that is used consists of six spectral matrices corresponding to six time-frequency points of an ELF hiss-event spectrogram. The WDF analysis is performed on these six points and the results are compared with those obtained previously. A statistical stability analysis confirms the stability of the solutions. The WDF computation is fast and without any prespecified parameters. The regularization parameter has been chosen in accordance with the Morozov's discrepancy principle. The Generalized Cross Validation and L-curve criterions are then tentatively used to provide a fully data-driven method. However, these criterions fail to determine a suitable value of the regularization parameter. Although the entropy regularization leads to solutions that agree fairly well with those already published, some differences are observed, and these are discussed in detail. The main advantage of the ERA is to return the WDF that exhibits the largest entropy and to avoid the use of a priori models, which sometimes seem to be more accurate but without any justification.
Sandhu, Ali Imran; Desmal, Abdulla; Bagci, Hakan
2016-01-01
A sparsity-regularized Born iterative method (BIM) is proposed for efficiently reconstructing two-dimensional piecewise-continuous inhomogeneous dielectric profiles. Such profiles are typically not spatially sparse, which reduces the efficiency of the sparsity-promoting regularization. To overcome this problem, scattered fields are represented in terms of the spatial derivative of the dielectric profile and reconstruction is carried out over samples of the dielectric profile's derivative. Then, like the conventional BIM, the nonlinear problem is iteratively converted into a sequence of linear problems (in derivative samples) and sparsity constraint is enforced on each linear problem using the thresholded Landweber iterations. Numerical results, which demonstrate the efficiency and accuracy of the proposed method in reconstructing piecewise-continuous dielectric profiles, are presented.
Sandhu, Ali Imran
2016-04-10
A sparsity-regularized Born iterative method (BIM) is proposed for efficiently reconstructing two-dimensional piecewise-continuous inhomogeneous dielectric profiles. Such profiles are typically not spatially sparse, which reduces the efficiency of the sparsity-promoting regularization. To overcome this problem, scattered fields are represented in terms of the spatial derivative of the dielectric profile and reconstruction is carried out over samples of the dielectric profile\\'s derivative. Then, like the conventional BIM, the nonlinear problem is iteratively converted into a sequence of linear problems (in derivative samples) and sparsity constraint is enforced on each linear problem using the thresholded Landweber iterations. Numerical results, which demonstrate the efficiency and accuracy of the proposed method in reconstructing piecewise-continuous dielectric profiles, are presented.
Sun, Qi; Fu, Shujun
2017-09-20
Fringe orientation is an important feature of fringe patterns and has a wide range of applications such as guiding fringe pattern filtering, phase unwrapping, and abstraction. Estimating fringe orientation is a basic task for subsequent processing of fringe patterns. However, various noise, singular and obscure points, and orientation data degeneration lead to inaccurate calculations of fringe orientation. Thus, to deepen the understanding of orientation estimation and to better guide orientation estimation in fringe pattern processing, some advanced gradient-field-based orientation estimation methods are compared and analyzed. At the same time, following the ideas of smoothing regularization and computing of bigger gradient fields, a regularized singular-value decomposition (RSVD) technique is proposed for fringe orientation estimation. To compare the performance of these gradient-field-based methods, quantitative results and visual effect maps of orientation estimation are given on simulated and real fringe patterns that demonstrate that the RSVD produces the best estimation results at a cost of relatively less time.
Exact Heat Kernel on a Hypersphere and Its Applications in Kernel SVM
Directory of Open Access Journals (Sweden)
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.
Lestari, D.; Raharjo, D.; Bustamam, A.; Abdillah, B.; Widhianto, W.
2017-07-01
Dengue virus consists of 10 different constituent proteins and are classified into 4 major serotypes (DEN 1 - DEN 4). This study was designed to perform clustering against 30 protein sequences of dengue virus taken from Virus Pathogen Database and Analysis Resource (VIPR) using Regularized Markov Clustering (R-MCL) algorithm and then we analyze the result. By using Python program 3.4, R-MCL algorithm produces 8 clusters with more than one centroid in several clusters. The number of centroid shows the density level of interaction. Protein interactions that are connected in a tissue, form a complex protein that serves as a specific biological process unit. The analysis of result shows the R-MCL clustering produces clusters of dengue virus family based on the similarity role of their constituent protein, regardless of serotypes.
Chen, Tai-Been; Chen, Jyh-Cheng; Lu, Henry Horng-Shing
2012-01-01
Segmentation of positron emission tomography (PET) is typically achieved using the K-Means method or other approaches. In preclinical and clinical applications, the K-Means method needs a prior estimation of parameters such as the number of clusters and appropriate initialized values. This work segments microPET images using a hybrid method combining the Gaussian mixture model (GMM) with kernel density estimation. Segmentation is crucial to registration of disordered 2-deoxy-2-fluoro-D-glucose (FDG) accumulation locations with functional diagnosis and to estimate standardized uptake values (SUVs) of region of interests (ROIs) in PET images. Therefore, simulation studies are conducted to apply spherical targets to evaluate segmentation accuracy based on Tanimoto's definition of similarity. The proposed method generates a higher degree of similarity than the K-Means method. The PET images of a rat brain are used to compare the segmented shape and area of the cerebral cortex by the K-Means method and the proposed method by volume rendering. The proposed method provides clearer and more detailed activity structures of an FDG accumulation location in the cerebral cortex than those by the K-Means method.
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.
International Nuclear Information System (INIS)
Itagaki, Masafumi; Miyoshi, Yoshinori; Hirose, Hideyuki
1993-01-01
A procedure is presented for the determination of geometric buckling for regular polygons. A new computation technique, the multiple reciprocity boundary element method (MRBEM), has been applied to solve the one-group neutron diffusion equation. The main difficulty in applying the ordinary boundary element method (BEM) to neutron diffusion problems has been the need to compute a domain integral, resulting from the fission source. The MRBEM has been developed for transforming this type of domain integral into an equivalent boundary integral. The basic idea of the MRBEM is to apply repeatedly the reciprocity theorem (Green's second formula) using a sequence of higher order fundamental solutions. The MRBEM requires discretization of the boundary only rather than of the domain. This advantage is useful for extensive survey analyses of buckling for complex geometries. The results of survey analyses have indicated that the general form of geometric buckling is B g 2 = (a n /R c ) 2 , where R c represents the radius of the circumscribed circle of the regular polygon under consideration. The geometric constant A n depends on the type of regular polygon and takes the value of π for a square and 2.405 for a circle, an extreme case that has an infinite number of sides. Values of a n for a triangle, pentagon, hexagon, and octagon have been calculated as 4.190, 2.281, 2.675, and 2.547, respectively
Directory of Open Access Journals (Sweden)
Shuang Liu
2018-01-01
Full Text Available In this paper, the eigenmode linear superposition (ELS method based on the regularization is used to discuss the distributions of all eigenmodes and the role of their instability to the intensity and structure change in TC-like vortex. Results show that the regularization approach can overcome the ill-posed problem occurring in solving mode weight coefficients as the ELS method are applied to analyze the impacts of dynamic instability on the intensity and structure change of TC-like vortex. The Generalized Cross-validation (GCV method and the L curve method are used to determine the regularization parameters, and the results of the two approaches are compared. It is found that the results based on the GCV method are closer to the given initial condition in the solution of the inverse problem of the vortex system. Then, the instability characteristic of the hollow vortex as the basic state are examined based on the linear barotropic shallow water equations. It is shown that the wavenumber distribution of system instability obtained from the ELS method is well consistent with that of the numerical analysis based on the norm mode. On the other hand, the evolution of the hollow vortex are discussed using the product of each eigenmode and its corresponding weight coefficient. Results show that the intensity and structure change of the system are mainly affected by the dynamic instability in the early stage of disturbance development, and the most unstable mode has a dominant role in the growth rate and the horizontal distribution of intense disturbance in the near-core region. Moreover, the wave structure of the most unstable mode possesses typical characteristics of mixed vortex Rossby-inertio-gravity waves (VRIGWs.
3D Inversion of Magnetic Data through Wavelet based Regularization Method
Directory of Open Access Journals (Sweden)
Maysam Abedi
2015-06-01
Full Text Available This study deals with the 3D recovering of magnetic susceptibility model by incorporating the sparsity-based constraints in the inversion algorithm. For this purpose, the area under prospect was divided into a large number of rectangular prisms in a mesh with unknown susceptibilities. Tikhonov cost functions with two sparsity functions were used to recover the smooth parts as well as the sharp boundaries of model parameters. A pre-selected basis namely wavelet can recover the region of smooth behaviour of susceptibility distribution while Haar or finite-difference (FD domains yield a solution with rough boundaries. Therefore, a regularizer function which can benefit from the advantages of both wavelets and Haar/FD operators in representation of the 3D magnetic susceptibility distributionwas chosen as a candidate for modeling magnetic anomalies. The optimum wavelet and parameter β which controls the weight of the two sparsifying operators were also considered. The algorithm assumed that there was no remanent magnetization and observed that magnetometry data represent only induced magnetization effect. The proposed approach is applied to a noise-corrupted synthetic data in order to demonstrate its suitability for 3D inversion of magnetic data. On obtaining satisfactory results, a case study pertaining to the ground based measurement of magnetic anomaly over a porphyry-Cu deposit located in Kerman providence of Iran. Now Chun deposit was presented to be 3D inverted. The low susceptibility in the constructed model coincides with the known location of copper ore mineralization.
Noise reduction by support vector regression with a Ricker wavelet kernel
International Nuclear Information System (INIS)
Deng, Xiaoying; Yang, Dinghui; Xie, Jing
2009-01-01
We propose a noise filtering technology based on the least-squares support vector regression (LS-SVR), to improve the signal-to-noise ratio (SNR) of seismic data. We modified it by using an admissible support vector (SV) kernel, namely the Ricker wavelet kernel, to replace the conventional radial basis function (RBF) kernel in seismic data processing. We investigated the selection of the regularization parameter for the LS-SVR and derived a concise selecting formula directly from the noisy data. We used the proposed method for choosing the regularization parameter which not only had the advantage of high speed but could also obtain almost the same effectiveness as an optimal parameter method. We conducted experiments using synthetic data corrupted by the random noise of different types and levels, and found that our method was superior to the wavelet transform-based approach and the Wiener filtering. We also applied the method to two field seismic data sets and concluded that it was able to effectively suppress the random noise and improve the data quality in terms of SNR
Noise reduction by support vector regression with a Ricker wavelet kernel
Deng, Xiaoying; Yang, Dinghui; Xie, Jing
2009-06-01
We propose a noise filtering technology based on the least-squares support vector regression (LS-SVR), to improve the signal-to-noise ratio (SNR) of seismic data. We modified it by using an admissible support vector (SV) kernel, namely the Ricker wavelet kernel, to replace the conventional radial basis function (RBF) kernel in seismic data processing. We investigated the selection of the regularization parameter for the LS-SVR and derived a concise selecting formula directly from the noisy data. We used the proposed method for choosing the regularization parameter which not only had the advantage of high speed but could also obtain almost the same effectiveness as an optimal parameter method. We conducted experiments using synthetic data corrupted by the random noise of different types and levels, and found that our method was superior to the wavelet transform-based approach and the Wiener filtering. We also applied the method to two field seismic data sets and concluded that it was able to effectively suppress the random noise and improve the data quality in terms of SNR.
Directory of Open Access Journals (Sweden)
Özkan Güner
2014-01-01
Full Text Available We apply the functional variable method, exp-function method, and (G′/G-expansion method to establish the exact solutions of the nonlinear fractional partial differential equation (NLFPDE in the sense of the modified Riemann-Liouville derivative. As a result, some new exact solutions for them are obtained. The results show that these methods are very effective and powerful mathematical tools for solving nonlinear fractional equations arising in mathematical physics. As a result, these methods can also be applied to other nonlinear fractional differential equations.
Stochastic methods of data modeling: application to the reconstruction of non-regular data
International Nuclear Information System (INIS)
Buslig, Leticia
2014-01-01
This research thesis addresses two issues or applications related to IRSN studies. The first one deals with the mapping of measurement data (the IRSN must regularly control the radioactivity level in France and, for this purpose, uses a network of sensors distributed among the French territory). The objective is then to predict, by means of reconstruction model which used observations, maps which will be used to inform the population. The second application deals with the taking of uncertainties into account in complex computation codes (the IRSN must perform safety studies to assess the risks of loss of integrity of a nuclear reactor in case of hypothetical accidents, and for this purpose, codes are used which simulate physical phenomena occurring within an installation). Some input parameters are not precisely known, and the author therefore tries to assess the impact of some uncertainties on simulated values. She notably aims at seeing whether variations of input parameters may push the system towards a behaviour which is very different from that obtained with parameters having a reference value, or even towards a state in which safety conditions are not met. The precise objective of this second part is then to a reconstruction model which is not costly (in terms of computation time) and to perform simulation in relevant areas (strong gradient areas, threshold overrun areas, so on). Two issues are then important: the choice of the approximation model and the construction of the experiment plan. The model is based on a kriging-type stochastic approach, and an important part of the work addresses the development of new numerical techniques of experiment planning. The first part proposes a generic criterion of adaptive planning, and reports its analysis and implementation. In the second part, an alternative to error variance addition is developed. Methodological developments are tested on analytic functions, and then applied to the cases of measurement mapping and
Less is more: regularization perspectives on large scale machine learning
CERN. Geneva
2017-01-01
Deep learning based techniques provide a possible solution at the expanse of theoretical guidance and, especially, of computational requirements. It is then a key challenge for large scale machine learning to devise approaches guaranteed to be accurate and yet computationally efficient. In this talk, we will consider a regularization perspectives on machine learning appealing to classical ideas in linear algebra and inverse problems to scale-up dramatically nonparametric methods such as kernel methods, often dismissed because of prohibitive costs. Our analysis derives optimal theoretical guarantees while providing experimental results at par or out-performing state of the art approaches.
Heat kernels and zeta functions on fractals
International Nuclear Information System (INIS)
Dunne, Gerald V
2012-01-01
On fractals, spectral functions such as heat kernels and zeta functions exhibit novel features, very different from their behaviour on regular smooth manifolds, and these can have important physical consequences for both classical and quantum physics in systems having fractal properties. This article is part of a special issue of Journal of Physics A: Mathematical and Theoretical in honour of Stuart Dowker's 75th birthday devoted to ‘Applications of zeta functions and other spectral functions in mathematics and physics’. (paper)
Regularization of DT-MR images using a successive Fermat median filtering method.
Kwon, Kiwoon; Kim, Dongyoun; Kim, Sunghee; Park, Insung; Jeong, Jaewon; Kim, Taehwan; Hong, Cheolpyo; Han, Bongsoo
2008-05-21
Tractography using diffusion tensor magnetic resonance imaging (DT-MRI) is a method to determine the architecture of axonal fibers in the central nervous system by computing the direction of greatest diffusion in the white matter of the brain. To reduce the noise in DT-MRI measurements, a tensor-valued median filter, which is reported to be denoising and structure preserving in the tractography, is applied. In this paper, we proposed the successive Fermat (SF) method, successively using Fermat point theory for a triangle contained in the two-dimensional plane, as a median filtering method. We discussed the error analysis and numerical study about the SF method for phantom and experimental data. By considering the computing time and the image quality aspects of the numerical study simultaneously, we showed that the SF method is much more efficient than the simple median (SM) and gradient descents (GD) methods.
Regularization of DT-MR images using a successive Fermat median filtering method
International Nuclear Information System (INIS)
Kwon, Kiwoon; Kim, Dongyoun; Kim, Sunghee; Park, Insung; Jeong, Jaewon; Kim, Taehwan; Hong, Cheolpyo; Han, Bongsoo
2008-01-01
Tractography using diffusion tensor magnetic resonance imaging (DT-MRI) is a method to determine the architecture of axonal fibers in the central nervous system by computing the direction of greatest diffusion in the white matter of the brain. To reduce the noise in DT-MRI measurements, a tensor-valued median filter, which is reported to be denoising and structure preserving in the tractography, is applied. In this paper, we proposed the successive Fermat (SF) method, successively using Fermat point theory for a triangle contained in the two-dimensional plane, as a median filtering method. We discussed the error analysis and numerical study about the SF method for phantom and experimental data. By considering the computing time and the image quality aspects of the numerical study simultaneously, we showed that the SF method is much more efficient than the simple median (SM) and gradient descents (GD) methods
Regularization of DT-MR images using a successive Fermat median filtering method
Energy Technology Data Exchange (ETDEWEB)
Kwon, Kiwoon; Kim, Dongyoun; Kim, Sunghee; Park, Insung; Jeong, Jaewon; Kim, Taehwan [Department of Biomedical Engineering, Yonsei University, Wonju, 220-710 (Korea, Republic of); Hong, Cheolpyo; Han, Bongsoo [Department of Radiological Science, Yonsei University, Wonju, 220-710 (Korea, Republic of)], E-mail: bshan@yonsei.ac.kr
2008-05-21
Tractography using diffusion tensor magnetic resonance imaging (DT-MRI) is a method to determine the architecture of axonal fibers in the central nervous system by computing the direction of greatest diffusion in the white matter of the brain. To reduce the noise in DT-MRI measurements, a tensor-valued median filter, which is reported to be denoising and structure preserving in the tractography, is applied. In this paper, we proposed the successive Fermat (SF) method, successively using Fermat point theory for a triangle contained in the two-dimensional plane, as a median filtering method. We discussed the error analysis and numerical study about the SF method for phantom and experimental data. By considering the computing time and the image quality aspects of the numerical study simultaneously, we showed that the SF method is much more efficient than the simple median (SM) and gradient descents (GD) methods.
Cockrell, C. R.
1989-01-01
Numerical solutions of the differential equation which describe the electric field within an inhomogeneous layer of permittivity, upon which a perpendicularly-polarized plane wave is incident, are considered. Richmond's method and the Runge-Kutta method are compared for linear and exponential profiles of permittivities. These two approximate solutions are also compared with the exact solutions.
Identification of Fusarium damaged wheat kernels using image analysis
Directory of Open Access Journals (Sweden)
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.
Energy Technology Data Exchange (ETDEWEB)
Mory, Cyril, E-mail: cyril.mory@philips.com [Université de Lyon, CREATIS, CNRS UMR5220, Inserm U1044, INSA-Lyon, Université Lyon 1, F-69621 Villeurbanne Cedex (France); Philips Research Medisys, 33 rue de Verdun, 92156 Suresnes (France); Auvray, Vincent; Zhang, Bo [Philips Research Medisys, 33 rue de Verdun, 92156 Suresnes (France); Grass, Michael; Schäfer, Dirk [Philips Research, Röntgenstrasse 24–26, D-22335 Hamburg (Germany); Chen, S. James; Carroll, John D. [Department of Medicine, Division of Cardiology, University of Colorado Denver, 12605 East 16th Avenue, Aurora, Colorado 80045 (United States); Rit, Simon [Université de Lyon, CREATIS, CNRS UMR5220, Inserm U1044, INSA-Lyon, Université Lyon 1 (France); Centre Léon Bérard, 28 rue Laënnec, F-69373 Lyon (France); Peyrin, Françoise [Université de Lyon, CREATIS, CNRS UMR5220, Inserm U1044, INSA-Lyon, Université Lyon 1, F-69621 Villeurbanne Cedex (France); X-ray Imaging Group, European Synchrotron, Radiation Facility, BP 220, F-38043 Grenoble Cedex (France); Douek, Philippe; Boussel, Loïc [Université de Lyon, CREATIS, CNRS UMR5220, Inserm U1044, INSA-Lyon, Université Lyon 1 (France); Hospices Civils de Lyon, 28 Avenue du Doyen Jean Lépine, 69500 Bron (France)
2014-02-15
Purpose: Reconstruction of the beating heart in 3D + time in the catheter laboratory using only the available C-arm system would improve diagnosis, guidance, device sizing, and outcome control for intracardiac interventions, e.g., electrophysiology, valvular disease treatment, structural or congenital heart disease. To obtain such a reconstruction, the patient's electrocardiogram (ECG) must be recorded during the acquisition and used in the reconstruction. In this paper, the authors present a 4D reconstruction method aiming to reconstruct the heart from a single sweep 10 s acquisition. Methods: The authors introduce the 4D RecOnstructiOn using Spatial and TEmporal Regularization (short 4D ROOSTER) method, which reconstructs all cardiac phases at once, as a 3D + time volume. The algorithm alternates between a reconstruction step based on conjugate gradient and four regularization steps: enforcing positivity, averaging along time outside a motion mask that contains the heart and vessels, 3D spatial total variation minimization, and 1D temporal total variation minimization. Results: 4D ROOSTER recovers the different temporal representations of a moving Shepp and Logan phantom, and outperforms both ECG-gated simultaneous algebraic reconstruction technique and prior image constrained compressed sensing on a clinical case. It generates 3D + time reconstructions with sharp edges which can be used, for example, to estimate the patient's left ventricular ejection fraction. Conclusions: 4D ROOSTER can be applied for human cardiac C-arm CT, and potentially in other dynamic tomography areas. It can easily be adapted to other problems as regularization is decoupled from projection and back projection.
International Nuclear Information System (INIS)
Mory, Cyril; Auvray, Vincent; Zhang, Bo; Grass, Michael; Schäfer, Dirk; Chen, S. James; Carroll, John D.; Rit, Simon; Peyrin, Françoise; Douek, Philippe; Boussel, Loïc
2014-01-01
Purpose: Reconstruction of the beating heart in 3D + time in the catheter laboratory using only the available C-arm system would improve diagnosis, guidance, device sizing, and outcome control for intracardiac interventions, e.g., electrophysiology, valvular disease treatment, structural or congenital heart disease. To obtain such a reconstruction, the patient's electrocardiogram (ECG) must be recorded during the acquisition and used in the reconstruction. In this paper, the authors present a 4D reconstruction method aiming to reconstruct the heart from a single sweep 10 s acquisition. Methods: The authors introduce the 4D RecOnstructiOn using Spatial and TEmporal Regularization (short 4D ROOSTER) method, which reconstructs all cardiac phases at once, as a 3D + time volume. The algorithm alternates between a reconstruction step based on conjugate gradient and four regularization steps: enforcing positivity, averaging along time outside a motion mask that contains the heart and vessels, 3D spatial total variation minimization, and 1D temporal total variation minimization. Results: 4D ROOSTER recovers the different temporal representations of a moving Shepp and Logan phantom, and outperforms both ECG-gated simultaneous algebraic reconstruction technique and prior image constrained compressed sensing on a clinical case. It generates 3D + time reconstructions with sharp edges which can be used, for example, to estimate the patient's left ventricular ejection fraction. Conclusions: 4D ROOSTER can be applied for human cardiac C-arm CT, and potentially in other dynamic tomography areas. It can easily be adapted to other problems as regularization is decoupled from projection and back projection
Chen, Tianle; Zeng, Donglin
2015-01-01
Summary Predicting disease risk and progression is one of the main goals in many clinical research studies. Cohort studies on the natural history and etiology of chronic diseases span years and data are collected at multiple visits. Although kernel-based statistical learning methods are proven to be powerful for a wide range of disease prediction problems, these methods are only well studied for independent data but not for longitudinal data. It is thus important to develop time-sensitive prediction rules that make use of the longitudinal nature of the data. In this paper, we develop a novel statistical learning method for longitudinal data by introducing subject-specific short-term and long-term latent effects through a designed kernel to account for within-subject correlation of longitudinal measurements. Since the presence of multiple sources of data is increasingly common, we embed our method in a multiple kernel learning framework and propose a regularized multiple kernel statistical learning with random effects to construct effective nonparametric prediction rules. Our method allows easy integration of various heterogeneous data sources and takes advantage of correlation among longitudinal measures to increase prediction power. We use different kernels for each data source taking advantage of the distinctive feature of each data modality, and then optimally combine data across modalities. We apply the developed methods to two large epidemiological studies, one on Huntington's disease and the other on Alzheimer's Disease (Alzheimer's Disease Neuroimaging Initiative, ADNI) where we explore a unique opportunity to combine imaging and genetic data to study prediction of mild cognitive impairment, and show a substantial gain in performance while accounting for the longitudinal aspect of the data. PMID:26177419
Directory of Open Access Journals (Sweden)
Bérenger A. L. Ladele
2017-10-01
Full Text Available The comparative study of physico-chemical characteristics of Terminalia catappa L. kernel oils extracted by two methods has been done. The oil yields were 28.13 % and 61.78 % respectively for traditional and hexane methods and the fatty acid profiles showed palmitic acid (40.79 % and 40.03 % respectively oleic acid (25.55 % and 26.09 % respectively, linoleic acid (26.72 % and 26.64 % respectively and stearic acid (4.35 % and 4.49 % respectively as major components. The oils extracted by the two ways showed similar physico-chemical properties, good calorific values and non-toxicity against Artemia salina L. Oil obtained by traditional method exhibited more antioxidant capacity (1.40 than the hexane one (0.15. This traditional method helps to extract 45 % of the total oil. It gives oil free of organic solvent, with good physico-chemical properties that could be useful as edible oil and for industrial applications.
Žvokelj, Matej; Zupan, Samo; Prebil, Ivan
2011-10-01
The article presents a novel non-linear multivariate and multiscale statistical process monitoring and signal denoising method which combines the strengths of the Kernel Principal Component Analysis (KPCA) non-linear multivariate monitoring approach with the benefits of Ensemble Empirical Mode Decomposition (EEMD) to handle multiscale system dynamics. The proposed method which enables us to cope with complex even severe non-linear systems with a wide dynamic range was named the EEMD-based multiscale KPCA (EEMD-MSKPCA). The method is quite general in nature and could be used in different areas for various tasks even without any really deep understanding of the nature of the system under consideration. Its efficiency was first demonstrated by an illustrative example, after which the applicability for the task of bearing fault detection, diagnosis and signal denosing was tested on simulated as well as actual vibration and acoustic emission (AE) signals measured on purpose-built large-size low-speed bearing test stand. The positive results obtained indicate that the proposed EEMD-MSKPCA method provides a promising tool for tackling non-linear multiscale data which present a convolved picture of many events occupying different regions in the time-frequency plane.
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
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.
DEFF Research Database (Denmark)
Hejlesen, Mads Mølholm; Spietz, Henrik J.; Walther, Jens Honore
2014-01-01
, unbounded particle-mesh based vortex method is used to simulate the instability, transition to turbulence and eventual destruction of a single vortex ring. From the simulation data a novel method on analyzing the dynamics of the enstrophy is presented based on the alignment of the vorticity vector...... with the principal axis of the strain rate tensor. We find that the dynamics of the enstrophy density is dominated by the local flow deformation and axis of rotation, which is used to infer some concrete tendencies related to the topology of the vorticity field....
CompareSVM: supervised, Support Vector Machine (SVM) inference of gene regularity networks.
Gillani, Zeeshan; Akash, Muhammad Sajid Hamid; Rahaman, M D Matiur; Chen, Ming
2014-11-30
Predication of gene regularity network (GRN) from expression data is a challenging task. There are many methods that have been developed to address this challenge ranging from supervised to unsupervised methods. Most promising methods are based on support vector machine (SVM). There is a need for comprehensive analysis on prediction accuracy of supervised method SVM using different kernels on different biological experimental conditions and network size. We developed a tool (CompareSVM) based on SVM to compare different kernel methods for inference of GRN. Using CompareSVM, we investigated and evaluated different SVM kernel methods on simulated datasets of microarray of different sizes in detail. The results obtained from CompareSVM showed that accuracy of inference method depends upon the nature of experimental condition and size of the network. For network with nodes (SVM Gaussian kernel outperform on knockout, knockdown, and multifactorial datasets compared to all the other inference methods. For network with large number of nodes (~500), choice of inference method depend upon nature of experimental condition. CompareSVM is available at http://bis.zju.edu.cn/CompareSVM/ .
International Nuclear Information System (INIS)
Dmitriy Y. Anistratov; Adrian Constantinescu; Loren Roberts; William Wieselquist
2007-01-01
This is a project in the field of fundamental research on numerical methods for solving the particle transport equation. Numerous practical problems require to use unstructured meshes, for example, detailed nuclear reactor assembly-level calculations, large-scale reactor core calculations, radiative hydrodynamics problems, where the mesh is determined by hydrodynamic processes, and well-logging problems in which the media structure has very complicated geometry. Currently this is an area of very active research in numerical transport theory. main issues in developing numerical methods for solving the transport equation are the accuracy of the numerical solution and effectiveness of iteration procedure. The problem in case of unstructured grids is that it is very difficult to derive an iteration algorithm that will be unconditionally stable
A comparative analysis of the EEDF obtained by Regularization and by Least square fit methods
International Nuclear Information System (INIS)
Gutierrez T, C.; Flores Ll, H.
2004-01-01
The second derived of the characteristic curve current-voltage (I - V) of a Langmuir probe (I - V) is numerically calculated using the Tikhonov method for to determine the distribution function of the electrons energy (EEDF). One comparison of the obtained EEDF and a fit by least square are discussed (LS). The I - V experimental curve is obtained in a plasma source in the electron cyclotron resonance (ECR) using a cylindrical probe. The parameters of plasma are determined of the EEDF by means of the Laframboise theory. For the case of the LS fit, the obtained results are similar to those obtained by the Tikhonov method, but in the first case the procedure is slow to achieve the best fit. (Author)
International Nuclear Information System (INIS)
Soussaline, F.; LeCoq, C.; Raynaud, C.; Kellershohn, C.
1982-09-01
The aim of this study is to evaluate the potential of the RIM technique when used in brain studies. The analytical Regulatorizing Iterative Method (RIM) is designed to provide fast and accurate reconstruction of tomographic images when non-uniform attenuation is to be accounted for. As indicated by phantom studies, this method improves the contrast and the signal-to-noise ratio as compared to those obtained with FBP (Filtered Back Projection) technique. Preliminary results obtained in brain studies using AMPI-123 (isopropil-amphetamine I-123) are very encouraging in terms of quantitative regional cellular activity. However, the clinical usefulness of this mathematically accurate reconstruction procedure is going to be demonstrated in our Institution, in comparing quantitative data in heart or liver studies where control values can be obtained
Kursun, Zerrin; Cali, Sanda; Sakarya, Sibel
2014-06-01
To evaluate the demand, efficacy, and satisfaction concerning the Standard Days Method(®) (SDM; a fertility awareness method) as an option presented among other contraceptive methods at regular service delivery settings. The survey group consisted of 993 women who presented at the primary care units in Umraniye District of Istanbul, Turkey, between 1 October 2006 and 31 March 2008, and started to use a new method. Women were enrolled until reaching a limit of 250 new users for each method, or expiration of the six-month registration period. Participants were followed for up to one year of method use. The characteristics of women who chose the SDM were similar to those of participants who opted for other methods. The most common reasons for selecting it were that it is natural and causes no side effects. Fifty-one percent used the SDM for the full year, compared to 71% who chose an intrauterine device (IUD). Continuation rates were significantly lower for all other methods. During the one-year follow-up period, 12% of SDM-, 7% of pill-, 7% of condom-, 3% of monthly injection-, 1% of quarterly injection-, and 0.5% of IUD users became pregnant. The SDM had relatively high continuation rates and relatively good levels of satisfaction among participants and their husbands. It should be mentioned among the routinely offered contraceptive methods.
Directory of Open Access Journals (Sweden)
Kuzmichev Andrey A.
2017-01-01
Full Text Available Due to the active step of urbanization and rapid development of industry the external appearance of buildings and architectural monuments of urban environment from visual ecology position requires special attention. Dust deposition by polluted atmospheric air is one of the key aspects of degradation of the facades of buildings. With the help of modern computer modeling methods it is possible to evaluate the impact of polluted atmospheric air on the external facades of the buildings in order to save them.
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.
Feature selection and multi-kernel learning for sparse representation on a manifold
Wang, Jim Jing-Yan
2014-03-01
Sparse representation has been widely studied as a part-based data representation method and applied in many scientific and engineering fields, such as bioinformatics and medical imaging. It seeks to represent a data sample as a sparse linear combination of some basic items in a dictionary. Gao etal. (2013) recently proposed Laplacian sparse coding by regularizing the sparse codes with an affinity graph. However, due to the noisy features and nonlinear distribution of the data samples, the affinity graph constructed directly from the original feature space is not necessarily a reliable reflection of the intrinsic manifold of the data samples. To overcome this problem, we integrate feature selection and multiple kernel learning into the sparse coding on the manifold. To this end, unified objectives are defined for feature selection, multiple kernel learning, sparse coding, and graph regularization. By optimizing the objective functions iteratively, we develop novel data representation algorithms with feature selection and multiple kernel learning respectively. Experimental results on two challenging tasks, N-linked glycosylation prediction and mammogram retrieval, demonstrate that the proposed algorithms outperform the traditional sparse coding methods. © 2013 Elsevier Ltd.
Feature selection and multi-kernel learning for sparse representation on a manifold.
Wang, Jim Jing-Yan; Bensmail, Halima; Gao, Xin
2014-03-01
Sparse representation has been widely studied as a part-based data representation method and applied in many scientific and engineering fields, such as bioinformatics and medical imaging. It seeks to represent a data sample as a sparse linear combination of some basic items in a dictionary. Gao et al. (2013) recently proposed Laplacian sparse coding by regularizing the sparse codes with an affinity graph. However, due to the noisy features and nonlinear distribution of the data samples, the affinity graph constructed directly from the original feature space is not necessarily a reliable reflection of the intrinsic manifold of the data samples. To overcome this problem, we integrate feature selection and multiple kernel learning into the sparse coding on the manifold. To this end, unified objectives are defined for feature selection, multiple kernel learning, sparse coding, and graph regularization. By optimizing the objective functions iteratively, we develop novel data representation algorithms with feature selection and multiple kernel learning respectively. Experimental results on two challenging tasks, N-linked glycosylation prediction and mammogram retrieval, demonstrate that the proposed algorithms outperform the traditional sparse coding methods. Copyright © 2013 Elsevier Ltd. All rights reserved.
2015-06-01
implementation of the direct interaction called particle-to-particle kernel for a shared-memory single GPU device using the Compute Unified Device Architecture ...GPU-defined P2P kernel we developed using the Compute Unified Device Architecture (CUDA).9 A brief outline of the rest of this work follows. The...Employed The computing environment used for this work is a 64-node heterogeneous cluster consisting of 48 IBM dx360M4 nodes, each with one Intel Phi
Calculation of the time resolution of the J-PET tomograph using kernel density estimation
Raczyński, L.; Wiślicki, W.; Krzemień, W.; Kowalski, P.; Alfs, D.; Bednarski, T.; Białas, P.; Curceanu, C.; Czerwiński, E.; Dulski, K.; Gajos, A.; Głowacz, B.; Gorgol, M.; Hiesmayr, B.; Jasińska, B.; Kamińska, D.; Korcyl, G.; Kozik, T.; Krawczyk, N.; Kubicz, E.; Mohammed, M.; Pawlik-Niedźwiecka, M.; Niedźwiecki, S.; Pałka, M.; Rudy, Z.; Rundel, O.; Sharma, N. G.; Silarski, M.; Smyrski, J.; Strzelecki, A.; Wieczorek, A.; Zgardzińska, B.; Zieliński, M.; Moskal, P.
2017-06-01
In this paper we estimate the time resolution of the J-PET scanner built from plastic scintillators. We incorporate the method of signal processing using the Tikhonov regularization framework and the kernel density estimation method. We obtain simple, closed-form analytical formulae for time resolution. The proposed method is validated using signals registered by means of the single detection unit of the J-PET tomograph built from a 30 cm long plastic scintillator strip. It is shown that the experimental and theoretical results obtained for the J-PET scanner equipped with vacuum tube photomultipliers are consistent.
SFAK, Unscattered Gamma Self-Absorption from Regular Fuel Rod Assemblies
International Nuclear Information System (INIS)
Wand, H.
1982-01-01
1 - Description of problem or function: Calculation of the self- absorption of unscattered (gamma-) radiation from fuel assemblies which contain a regular arrangement of identical fuel rods. 2 - Method of solution: The point-kernel is integrated over the radiation sources, i.e. the fuel rods. A uniform mesh of integration points is used for each of the fuel rods. 3 - Restrictions on the complexity of the problem: Number of fuel rods is dynamically allocated
Intelligent Control of a Sensor-Actuator System via Kernelized Least-Squares Policy Iteration
Directory of Open Access Journals (Sweden)
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.
Multi-task feature learning by using trace norm regularization
Directory of Open Access Journals (Sweden)
Jiangmei Zhang
2017-11-01
Full Text Available Multi-task learning can extract the correlation of multiple related machine learning problems to improve performance. This paper considers applying the multi-task learning method to learn a single task. We propose a new learning approach, which employs the mixture of expert model to divide a learning task into several related sub-tasks, and then uses the trace norm regularization to extract common feature representation of these sub-tasks. A nonlinear extension of this approach by using kernel is also provided. Experiments conducted on both simulated and real data sets demonstrate the advantage of the proposed approach.
Hopfer, Helene; Jodari, Farman; Negre-Zakharov, Florence; Wylie, Phillip L; Ebeler, Susan E
2016-05-25
Demand for aromatic rice varieties (e.g., Basmati) is increasing in the US. Aromatic varieties typically have elevated levels of the aroma compound 2-acetyl-1-pyrroline (2AP). Due to its very low aroma threshold, analysis of 2AP provides a useful screening tool for rice breeders. Methods for 2AP analysis in rice should quantitate 2AP at or below sensory threshold level, avoid artifactual 2AP generation, and be able to analyze single rice kernels in cases where only small sample quantities are available (e.g., breeding trials). We combined headspace solid phase microextraction with gas chromatography tandem mass spectrometry (HS-SPME-GC-MS/MS) for analysis of 2AP, using an extraction temperature of 40 °C and a stable isotopologue as internal standard. 2AP calibrations were linear between the concentrations of 53 and 5380 pg/g, with detection limits below the sensory threshold of 2AP. Forty-eight aromatic and nonaromatic, milled rice samples from three harvest years were screened with the method for their 2AP content, and overall reproducibility, observed for all samples, ranged from 5% for experimental aromatic lines to 33% for nonaromatic lines.
Classification of maize kernels using NIR hyperspectral imaging
DEFF Research Database (Denmark)
Williams, Paul; Kucheryavskiy, Sergey V.
2016-01-01
NIR hyperspectral imaging was evaluated to classify maize kernels of three hardness categories: hard, medium and soft. Two approaches, pixel-wise and object-wise, were investigated to group kernels according to hardness. The pixel-wise classification assigned a class to every pixel from individual...... and specificity of 0.95 and 0.93). Both feature extraction methods can be recommended for classification of maize kernels on production scale....
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.
DEFF Research Database (Denmark)
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...
Adaptive metric kernel regression
DEFF Research Database (Denmark)
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
DEFF Research Database (Denmark)
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...
Energy Technology Data Exchange (ETDEWEB)
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.
International Nuclear Information System (INIS)
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.
Dose point kernels for beta-emitting radioisotopes
International Nuclear Information System (INIS)
Prestwich, W.V.; Chan, L.B.; Kwok, C.S.; Wilson, B.
1986-01-01
Knowledge of the dose point kernel corresponding to a specific radionuclide is required to calculate the spatial dose distribution produced in a homogeneous medium by a distributed source. Dose point kernels for commonly used radionuclides have been calculated previously using as a basis monoenergetic dose point kernels derived by numerical integration of a model transport equation. The treatment neglects fluctuations in energy deposition, an effect which has been later incorporated in dose point kernels calculated using Monte Carlo methods. This work describes new calculations of dose point kernels using the Monte Carlo results as a basis. An analytic representation of the monoenergetic dose point kernels has been developed. This provides a convenient method both for calculating the dose point kernel associated with a given beta spectrum and for incorporating the effect of internal conversion. An algebraic expression for allowed beta spectra has been accomplished through an extension of the Bethe-Bacher approximation, and tested against the exact expression. Simplified expression for first-forbidden shape factors have also been developed. A comparison of the calculated dose point kernel for 32 P with experimental data indicates good agreement with a significant improvement over the earlier results in this respect. An analytic representation of the dose point kernel associated with the spectrum of a single beta group has been formulated. 9 references, 16 figures, 3 tables
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.
Moghadam, Maryam Khazaee; Asl, Alireza Kamali; Geramifar, Parham; Zaidi, Habib
2016-01-01
Purpose: The aim of this work is to evaluate the application of tissue-specific dose kernels instead of water dose kernels to improve the accuracy of patient-specific dosimetry by taking tissue heterogeneities into consideration. Materials and Methods: Tissue-specific dose point kernels (DPKs) and
DEFF Research Database (Denmark)
Hansen, Lars Kai; Rasmussen, Carl Edward; Svarer, C.
1994-01-01
Regularization, e.g., in the form of weight decay, is important for training and optimization of neural network architectures. In this work the authors provide a tool based on asymptotic sampling theory, for iterative estimation of weight decay parameters. The basic idea is to do a gradient desce...
Collision kernels in the eikonal approximation for Lennard-Jones interaction potential
International Nuclear Information System (INIS)
Zielinska, S.
1985-03-01
The velocity changing collisions are conveniently described by collisional kernels. These kernels depend on an interaction potential and there is a necessity for evaluating them for realistic interatomic potentials. Using the collision kernels, we are able to investigate the redistribution of atomic population's caused by the laser light and velocity changing collisions. In this paper we present the method of evaluating the collision kernels in the eikonal approximation. We discuss the influence of the potential parameters Rsub(o)sup(i), epsilonsub(o)sup(i) on kernel width for a given atomic state. It turns out that unlike the collision kernel for the hard sphere model of scattering the Lennard-Jones kernel is not so sensitive to changes of Rsub(o)sup(i) as the previous one. Contrary to the general tendency of approximating collisional kernels by the Gaussian curve, kernels for the Lennard-Jones potential do not exhibit such a behaviour. (author)
DEFF Research Database (Denmark)
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...
DEFF Research Database (Denmark)
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.
Improved resolution and reliability in dynamic PET using Bayesian regularization of MRTM2
DEFF Research Database (Denmark)
Agn, Mikael; Svarer, Claus; Frokjaer, Vibe G.
2014-01-01
This paper presents a mathematical model that regularizes dynamic PET data by using a Bayesian framework. We base the model on the well known two-parameter multilinear reference tissue method MRTM2 and regularize on the assumption that spatially close regions have similar parameters. The developed...... model is compared to the conventional approach of improving the low signal-to-noise ratio of PET data, i.e., spatial filtering of each time frame independently by a Gaussian kernel. We show that the model handles high levels of noise better than the conventional approach, while at the same time...
Wilson Dslash Kernel From Lattice QCD Optimization
Energy Technology Data Exchange (ETDEWEB)
Joo, Balint [Jefferson Lab, Newport News, VA; Smelyanskiy, Mikhail [Parallel Computing Lab, Intel Corporation, California, USA; Kalamkar, Dhiraj D. [Parallel Computing Lab, Intel Corporation, India; Vaidyanathan, Karthikeyan [Parallel Computing Lab, Intel Corporation, India
2015-07-01
Lattice Quantum Chromodynamics (LQCD) is a numerical technique used for calculations in Theoretical Nuclear and High Energy Physics. LQCD is traditionally one of the first applications ported to many new high performance computing architectures and indeed LQCD practitioners have been known to design and build custom LQCD computers. Lattice QCD kernels are frequently used as benchmarks (e.g. 168.wupwise in the SPEC suite) and are generally well understood, and as such are ideal to illustrate several optimization techniques. In this chapter we will detail our work in optimizing the Wilson-Dslash kernels for Intel Xeon Phi, however, as we will show the technique gives excellent performance on regular Xeon Architecture as well.
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
Viscosity kernel of molecular fluids
DEFF Research Database (Denmark)
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...
Kernel Function Tuning for Single-Layer Neural Networks
Czech Academy of Sciences Publication Activity Database
Vidnerová, Petra; Neruda, Roman
-, accepted 28.11. 2017 (2018) ISSN 2278-0149 R&D Projects: GA ČR GA15-18108S Institutional support: RVO:67985807 Keywords : single-layer neural networks * kernel methods * kernel function * optimisation Subject RIV: IN - Informatics, Computer Science http://www.ijmerr.com/
Real time kernel performance monitoring with SystemTap
CERN. Geneva
2018-01-01
SystemTap is a dynamic method of monitoring and tracing the operation of a running Linux kernel. In this talk I will present a few practical use cases where SystemTap allowed me to turn otherwise complex userland monitoring tasks in simple kernel probes.
Optimal Bandwidth Selection in Observed-Score Kernel Equating
Häggström, Jenny; Wiberg, Marie
2014-01-01
The selection of bandwidth in kernel equating is important because it has a direct impact on the equated test scores. The aim of this article is to examine the use of double smoothing when selecting bandwidths in kernel equating and to compare double smoothing with the commonly used penalty method. This comparison was made using both an equivalent…
Directory of Open Access Journals (Sweden)
Chuang Lin
2015-01-01
Full Text Available Kernel Locality Preserving Projection (KLPP algorithm can effectively preserve the neighborhood structure of the database using the kernel trick. We have known that supervised KLPP (SKLPP can preserve within-class geometric structures by using label information. However, the conventional SKLPP algorithm endures the kernel selection which has significant impact on the performances of SKLPP. In order to overcome this limitation, a method named supervised kernel optimized LPP (SKOLPP is proposed in this paper, which can maximize the class separability in kernel learning. The proposed method maps the data from the original space to a higher dimensional kernel space using a data-dependent kernel. The adaptive parameters of the data-dependent kernel are automatically calculated through optimizing an objective function. Consequently, the nonlinear features extracted by SKOLPP have larger discriminative ability compared with SKLPP and are more adaptive to the input data. Experimental results on ORL, Yale, AR, and Palmprint databases showed the effectiveness of the proposed method.
Directory of Open Access Journals (Sweden)
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.
Czech Academy of Sciences Publication Activity Database
Branda, Martin; Bucher, M.; Červinka, Michal; Schwartz, A.
2018-01-01
Roč. 70, č. 2 (2018), s. 503-530 ISSN 0926-6003 R&D Projects: GA ČR GA15-00735S Institutional support: RVO:67985556 Keywords : Cardinality constraints * Regularization method * Scholtes regularization * Strong stationarity * Sparse portfolio optimization * Robust portfolio optimization Subject RIV: BB - Applied Statistics, Operational Research OBOR OECD: Statistics and probability Impact factor: 1.520, year: 2016 http://library.utia.cas.cz/separaty/2018/MTR/branda-0489264.pdf
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
Czech Academy of Sciences Publication Activity Database
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
Vested Madsen, Matias; Macario, Alex; Yamamoto, Satoshi; Tanaka, Pedro
2016-06-01
In this study, we examined the regularly scheduled, formal teaching sessions in a single anesthesiology residency program to (1) map the most common primary instructional methods, (2) map the use of 10 known teaching techniques, and (3) assess if residents scored sessions that incorporated active learning as higher quality than sessions with little or no verbal interaction between teacher and learner. A modified Delphi process was used to identify useful teaching techniques. A representative sample of each of the formal teaching session types was mapped, and residents anonymously completed a 5-question written survey rating the session. The most common primary instructional methods were computer slides-based classroom lectures (66%), workshops (15%), simulations (5%), and journal club (5%). The number of teaching techniques used per formal teaching session averaged 5.31 (SD, 1.92; median, 5; range, 0-9). Clinical applicability (85%) and attention grabbers (85%) were the 2 most common teaching techniques. Thirty-eight percent of the sessions defined learning objectives, and one-third of sessions engaged in active learning. The overall survey response rate equaled 42%, and passive sessions had a mean score of 8.44 (range, 5-10; median, 9; SD, 1.2) compared with a mean score of 8.63 (range, 5-10; median, 9; SD, 1.1) for active sessions (P = 0.63). Slides-based classroom lectures were the most common instructional method, and faculty used an average of 5 known teaching techniques per formal teaching session. The overall education scores of the sessions as rated by the residents were high.
Malas, Tareq Majed Yasin
2012-05-21
Several emerging petascale architectures use energy-efficient processors with vectorized computational units and in-order thread processing. On these architectures the sustained performance of streaming numerical kernels, ubiquitous in the solution of partial differential equations, represents a challenge despite the regularity of memory access. Sophisticated optimization techniques are required to fully utilize the CPU. We propose a new method for constructing streaming numerical kernels using a high-level assembly synthesis and optimization framework. We describe an implementation of this method in Python targeting the IBM® Blue Gene®/P supercomputer\\'s PowerPC® 450 core. This paper details the high-level design, construction, simulation, verification, and analysis of these kernels utilizing a subset of the CPU\\'s instruction set. We demonstrate the effectiveness of our approach by implementing several three-dimensional stencil kernels over a variety of cached memory scenarios and analyzing the mechanically scheduled variants, including a 27-point stencil achieving a 1.7× speedup over the best previously published results. © The Author(s) 2012.
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.
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
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 ...
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 .
An analysis of 1-D smoothed particle hydrodynamics kernels
International Nuclear Information System (INIS)
Fulk, D.A.; Quinn, D.W.
1996-01-01
In this paper, the smoothed particle hydrodynamics (SPH) kernel is analyzed, resulting in measures of merit for one-dimensional SPH. Various methods of obtaining an objective measure of the quality and accuracy of the SPH kernel are addressed. Since the kernel is the key element in the SPH methodology, this should be of primary concern to any user of SPH. The results of this work are two measures of merit, one for smooth data and one near shocks. The measure of merit for smooth data is shown to be quite accurate and a useful delineator of better and poorer kernels. The measure of merit for non-smooth data is not quite as accurate, but results indicate the kernel is much less important for these types of problems. In addition to the theory, 20 kernels are analyzed using the measure of merit demonstrating the general usefulness of the measure of merit and the individual kernels. In general, it was decided that bell-shaped kernels perform better than other shapes. 12 refs., 16 figs., 7 tabs
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...
Kernel based subspace projection of near infrared hyperspectral images of maize kernels
DEFF Research Database (Denmark)
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...
International Nuclear Information System (INIS)
Nam, H; Guo, M; Lee, K; Li, R; Xing, L; Gao, H
2014-01-01
Purpose: Inspired by compressive sensing, sparsity regularized iterative reconstruction method has been extensively studied. However, its utility pertinent to multislice helical 4D CT for radiotherapy with respect to imaging quality, dose, and time has not been thoroughly addressed. As the beginning of such an investigation, this work carries out the initial comparison of reconstructed imaging quality between sparsity regularized iterative method and analytic method through static phantom studies using a state-of-art 128-channel multi-slice Siemens helical CT scanner. Methods: In our iterative method, tensor framelet (TF) is chosen as the regularization method for its superior performance from total variation regularization in terms of reduced piecewise-constant artifacts and improved imaging quality that has been demonstrated in our prior work. On the other hand, X-ray transforms and its adjoints are computed on-the-fly through GPU implementation using our previous developed fast parallel algorithms with O(1) complexity per computing thread. For comparison, both FDK (approximate analytic method) and Katsevich algorithm (exact analytic method) are used for multislice helical CT image reconstruction. Results: The phantom experimental data with different imaging doses were acquired using a state-of-art 128-channel multi-slice Siemens helical CT scanner. The reconstructed image quality was compared between TF-based iterative method, FDK and Katsevich algorithm with the quantitative analysis for characterizing signal-to-noise ratio, image contrast, and spatial resolution of high-contrast and low-contrast objects. Conclusion: The experimental results suggest that our tensor framelet regularized iterative reconstruction algorithm improves the helical CT imaging quality from FDK and Katsevich algorithm for static experimental phantom studies that have been performed
DEFF Research Database (Denmark)
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...
Visualization of nonlinear kernel models in neuroimaging by sensitivity maps
DEFF Research Database (Denmark)
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...
Multiple kernel boosting framework based on information measure for classification
International Nuclear Information System (INIS)
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.
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.
Ingacheva, Anastasia; Chukalina, Marina; Khanipov, Timur; Nikolaev, Dmitry
2018-04-01
Motion blur caused by camera vibration is a common source of degradation in photographs. In this paper we study the problem of finding the point spread function (PSF) of a blurred image using the tomography technique. The PSF reconstruction result strongly depends on the particular tomography technique used. We present a tomography algorithm with regularization adapted specifically for this task. We use the algebraic reconstruction technique (ART algorithm) as the starting algorithm and introduce regularization. We use the conjugate gradient method for numerical implementation of the proposed approach. The algorithm is tested using a dataset which contains 9 kernels extracted from real photographs by the Adobe corporation where the point spread function is known. We also investigate influence of noise on the quality of image reconstruction and investigate how the number of projections influence the magnitude change of the reconstruction error.
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.
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...
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.
Explicit signal to noise ratio in reproducing kernel Hilbert spaces
DEFF Research Database (Denmark)
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...
Cid, Jaime A; von Davier, Alina A
2015-05-01
Test equating is a method of making the test scores from different test forms of the same assessment comparable. In the equating process, an important step involves continuizing the discrete score distributions. In traditional observed-score equating, this step is achieved using linear interpolation (or an unscaled uniform kernel). In the kernel equating (KE) process, this continuization process involves Gaussian kernel smoothing. It has been suggested that the choice of bandwidth in kernel smoothing controls the trade-off between variance and bias. In the literature on estimating density functions using kernels, it has also been suggested that the weight of the kernel depends on the sample size, and therefore, the resulting continuous distribution exhibits bias at the endpoints, where the samples are usually smaller. The purpose of this article is (a) to explore the potential effects of atypical scores (spikes) at the extreme ends (high and low) on the KE method in distributions with different degrees of asymmetry using the randomly equivalent groups equating design (Study I), and (b) to introduce the Epanechnikov and adaptive kernels as potential alternative approaches to reducing boundary bias in smoothing (Study II). The beta-binomial model is used to simulate observed scores reflecting a range of different skewed shapes.
Low-Complexity Regularization Algorithms for Image Deblurring
Alanazi, Abdulrahman
2016-11-01
Image restoration problems deal with images in which information has been degraded by blur or noise. In practice, the blur is usually caused by atmospheric turbulence, motion, camera shake, and several other mechanical or physical processes. In this study, we present two regularization algorithms for the image deblurring problem. We first present a new method based on solving a regularized least-squares (RLS) problem. This method is proposed to find a near-optimal value of the regularization parameter in the RLS problems. Experimental results on the non-blind image deblurring problem are presented. In all experiments, comparisons are made with three benchmark methods. The results demonstrate that the proposed method clearly outperforms the other methods in terms of both the output PSNR and structural similarity, as well as the visual quality of the deblurred images. To reduce the complexity of the proposed algorithm, we propose a technique based on the bootstrap method to estimate the regularization parameter in low and high-resolution images. Numerical results show that the proposed technique can effectively reduce the computational complexity of the proposed algorithms. In addition, for some cases where the point spread function (PSF) is separable, we propose using a Kronecker product so as to reduce the computations. Furthermore, in the case where the image is smooth, it is always desirable to replace the regularization term in the RLS problems by a total variation term. Therefore, we propose a novel method for adaptively selecting the regularization parameter in a so-called square root regularized total variation (SRTV). Experimental results demonstrate that our proposed method outperforms the other benchmark methods when applied to smooth images in terms of PSNR, SSIM and the restored image quality. In this thesis, we focus on the non-blind image deblurring problem, where the blur kernel is assumed to be known. However, we developed algorithms that also work
Stochastic subset selection for learning with kernel machines.
Rhinelander, Jason; Liu, Xiaoping P
2012-06-01
Kernel machines have gained much popularity in applications of machine learning. Support vector machines (SVMs) are a subset of kernel machines and generalize well for classification, regression, and anomaly detection tasks. The training procedure for traditional SVMs involves solving a quadratic programming (QP) problem. The QP problem scales super linearly in computational effort with the number of training samples and is often used for the offline batch processing of data. Kernel machines operate by retaining a subset of observed data during training. The data vectors contained within this subset are referred to as support vectors (SVs). The work presented in this paper introduces a subset selection method for the use of kernel machines in online, changing environments. Our algorithm works by using a stochastic indexing technique when selecting a subset of SVs when computing the kernel expansion. The work described here is novel because it separates the selection of kernel basis functions from the training algorithm used. The subset selection algorithm presented here can be used in conjunction with any online training technique. It is important for online kernel machines to be computationally efficient due to the real-time requirements of online environments. Our algorithm is an important contribution because it scales linearly with the number of training samples and is compatible with current training techniques. Our algorithm outperforms standard techniques in terms of computational efficiency and provides increased recognition accuracy in our experiments. We provide results from experiments using both simulated and real-world data sets to verify our algorithm.
Per-Sample Multiple Kernel Approach for Visual Concept Learning
Directory of Open Access Journals (Sweden)
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
Directory of Open Access Journals (Sweden)
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.
Deep Restricted Kernel Machines Using Conjugate Feature Duality.
Suykens, Johan A K
2017-08-01
The aim of this letter is to propose a theory of deep restricted kernel machines offering new foundations for deep learning with kernel machines. From the viewpoint of deep learning, it is partially related to restricted Boltzmann machines, which are characterized by visible and hidden units in a bipartite graph without hidden-to-hidden connections and deep learning extensions as deep belief networks and deep Boltzmann machines. From the viewpoint of kernel machines, it includes least squares support vector machines for classification and regression, kernel principal component analysis (PCA), matrix singular value decomposition, and Parzen-type models. A key element is to first characterize these kernel machines in terms of so-called conjugate feature duality, yielding a representation with visible and hidden units. It is shown how this is related to the energy form in restricted Boltzmann machines, with continuous variables in a nonprobabilistic setting. In this new framework of so-called restricted kernel machine (RKM) representations, the dual variables correspond to hidden features. Deep RKM are obtained by coupling the RKMs. The method is illustrated for deep RKM, consisting of three levels with a least squares support vector machine regression level and two kernel PCA levels. In its primal form also deep feedforward neural networks can be trained within this framework.
Selection of regularization parameter for l1-regularized damage detection
Hou, Rongrong; Xia, Yong; Bao, Yuequan; Zhou, Xiaoqing
2018-06-01
The l1 regularization technique has been developed for structural health monitoring and damage detection through employing the sparsity condition of structural damage. The regularization parameter, which controls the trade-off between data fidelity and solution size of the regularization problem, exerts a crucial effect on the solution. However, the l1 regularization problem has no closed-form solution, and the regularization parameter is usually selected by experience. This study proposes two strategies of selecting the regularization parameter for the l1-regularized damage detection problem. The first method utilizes the residual and solution norms of the optimization problem and ensures that they are both small. The other method is based on the discrepancy principle, which requires that the variance of the discrepancy between the calculated and measured responses is close to the variance of the measurement noise. The two methods are applied to a cantilever beam and a three-story frame. A range of the regularization parameter, rather than one single value, can be determined. When the regularization parameter in this range is selected, the damage can be accurately identified even for multiple damage scenarios. This range also indicates the sensitivity degree of the damage identification problem to the regularization parameter.
UNFOLDED REGULAR AND SEMI-REGULAR POLYHEDRA
Directory of Open Access Journals (Sweden)
IONIŢĂ Elena
2015-06-01
Full Text Available This paper proposes a presentation unfolding regular and semi-regular polyhedra. Regular polyhedra are convex polyhedra whose faces are regular and equal polygons, with the same number of sides, and whose polyhedral angles are also regular and equal. Semi-regular polyhedra are convex polyhedra with regular polygon faces, several types and equal solid angles of the same type. A net of a polyhedron is a collection of edges in the plane which are the unfolded edges of the solid. Modeling and unfolding Platonic and Arhimediene polyhedra will be using 3dsMAX program. This paper is intended as an example of descriptive geometry applications.
Insights from Classifying Visual Concepts with Multiple Kernel Learning
Binder, Alexander; Nakajima, Shinichi; Kloft, Marius; Müller, Christina; Samek, Wojciech; Brefeld, Ulf; Müller, Klaus-Robert; Kawanabe, Motoaki
2012-01-01
Combining information from various image features has become a standard technique in concept recognition tasks. However, the optimal way of fusing the resulting kernel functions is usually unknown in practical applications. Multiple kernel learning (MKL) techniques allow to determine an optimal linear combination of such similarity matrices. Classical approaches to MKL promote sparse mixtures. Unfortunately, 1-norm regularized MKL variants are often observed to be outperformed by an unweighted sum kernel. The main contributions of this paper are the following: we apply a recently developed non-sparse MKL variant to state-of-the-art concept recognition tasks from the application domain of computer vision. We provide insights on benefits and limits of non-sparse MKL and compare it against its direct competitors, the sum-kernel SVM and sparse MKL. We report empirical results for the PASCAL VOC 2009 Classification and ImageCLEF2010 Photo Annotation challenge data sets. Data sets (kernel matrices) as well as further information are available at http://doc.ml.tu-berlin.de/image_mkl/(Accessed 2012 Jun 25). PMID:22936970
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.
Manifold Regularized Reinforcement Learning.
Li, Hongliang; Liu, Derong; Wang, Ding
2018-04-01
This paper introduces a novel manifold regularized reinforcement learning scheme for continuous Markov decision processes. Smooth feature representations for value function approximation can be automatically learned using the unsupervised manifold regularization method. The learned features are data-driven, and can be adapted to the geometry of the state space. Furthermore, the scheme provides a direct basis representation extension for novel samples during policy learning and control. The performance of the proposed scheme is evaluated on two benchmark control tasks, i.e., the inverted pendulum and the energy storage problem. Simulation results illustrate the concepts of the proposed scheme and show that it can obtain excellent performance.
Chung, Moo K; Qiu, Anqi; Seo, Seongho; Vorperian, Houri K
2015-05-01
We present a novel kernel regression framework for smoothing scalar surface data using the Laplace-Beltrami eigenfunctions. Starting with the heat kernel constructed from the eigenfunctions, we formulate a new bivariate kernel regression framework as a weighted eigenfunction expansion with the heat kernel as the weights. The new kernel method is mathematically equivalent to isotropic heat diffusion, kernel smoothing and recently popular diffusion wavelets. The numerical implementation is validated on a unit sphere using spherical harmonics. As an illustration, the method is applied to characterize the localized growth pattern of mandible surfaces obtained in CT images between ages 0 and 20 by regressing the length of displacement vectors with respect to a surface template. Copyright © 2015 Elsevier B.V. All rights reserved.
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.
Fang, Jinsheng; Bao, Lijun; Li, Xu; van Zijl, Peter C. M.; Chen, Zhong
2017-08-01
Background field removal is an important MR phase preprocessing step for quantitative susceptibility mapping (QSM). It separates the local field induced by tissue magnetic susceptibility sources from the background field generated by sources outside a region of interest, e.g. brain, such as air-tissue interface. In the vicinity of air-tissue boundary, e.g. skull and paranasal sinuses, where large susceptibility variations exist, present background field removal methods are usually insufficient and these regions often need to be excluded by brain mask erosion at the expense of losing information of local field and thus susceptibility measures in these regions. In this paper, we propose an extension to the variable-kernel sophisticated harmonic artifact reduction for phase data (V-SHARP) background field removal method using a region adaptive kernel (R-SHARP), in which a scalable spherical Gaussian kernel (SGK) is employed with its kernel radius and weights adjustable according to an energy "functional" reflecting the magnitude of field variation. Such an energy functional is defined in terms of a contour and two fitting functions incorporating regularization terms, from which a curve evolution model in level set formation is derived for energy minimization. We utilize it to detect regions of with a large field gradient caused by strong susceptibility variation. In such regions, the SGK will have a small radius and high weight at the sphere center in a manner adaptive to the voxel energy of the field perturbation. Using the proposed method, the background field generated from external sources can be effectively removed to get a more accurate estimation of the local field and thus of the QSM dipole inversion to map local tissue susceptibility sources. Numerical simulation, phantom and in vivo human brain data demonstrate improved performance of R-SHARP compared to V-SHARP and RESHARP (regularization enabled SHARP) methods, even when the whole paranasal sinus regions
Fang, Jinsheng; Bao, Lijun; Li, Xu; van Zijl, Peter C M; Chen, Zhong
2017-08-01
Background field removal is an important MR phase preprocessing step for quantitative susceptibility mapping (QSM). It separates the local field induced by tissue magnetic susceptibility sources from the background field generated by sources outside a region of interest, e.g. brain, such as air-tissue interface. In the vicinity of air-tissue boundary, e.g. skull and paranasal sinuses, where large susceptibility variations exist, present background field removal methods are usually insufficient and these regions often need to be excluded by brain mask erosion at the expense of losing information of local field and thus susceptibility measures in these regions. In this paper, we propose an extension to the variable-kernel sophisticated harmonic artifact reduction for phase data (V-SHARP) background field removal method using a region adaptive kernel (R-SHARP), in which a scalable spherical Gaussian kernel (SGK) is employed with its kernel radius and weights adjustable according to an energy "functional" reflecting the magnitude of field variation. Such an energy functional is defined in terms of a contour and two fitting functions incorporating regularization terms, from which a curve evolution model in level set formation is derived for energy minimization. We utilize it to detect regions of with a large field gradient caused by strong susceptibility variation. In such regions, the SGK will have a small radius and high weight at the sphere center in a manner adaptive to the voxel energy of the field perturbation. Using the proposed method, the background field generated from external sources can be effectively removed to get a more accurate estimation of the local field and thus of the QSM dipole inversion to map local tissue susceptibility sources. Numerical simulation, phantom and in vivo human brain data demonstrate improved performance of R-SHARP compared to V-SHARP and RESHARP (regularization enabled SHARP) methods, even when the whole paranasal sinus regions
Energy Technology Data Exchange (ETDEWEB)
Baltazar R, A.; Vega C, H. R.; Ortiz R, J. M.; Solis S, L. O.; Castaneda M, R. [Universidad Autonoma de Zacatecas, Unidad Academica de Ingenieria Electrica, Programa de Doctorado en Ingenieria y Tecnologia Aplicada, Av. Lopez Velarde s/n, 98000 Zacatecas, Zac. (Mexico); Soto B, T. G.; Medina C, D., E-mail: raigosa.antonio@hotmail.com [Universidad Autonoma de Zacatecas, Unidad Academica de Estudios Nucleares, Programa de Doctorado en Ciencias Basicas (Ciencias Nucleares), Cipres No. 10, Fracc. La Penuela, 98060 Zacatecas, Zac. (Mexico)
2017-10-15
In the last three decades the uses of Monte Carlo methods, for the estimation of physical phenomena associated with the interaction of radiation with matter, have increased considerably. The reason is due to the increase in computing capabilities and the reduction of computer prices. Monte Carlo methods allow modeling and simulating real systems before their construction, saving time and costs. The interaction mechanisms between neutrons and matter are diverse and range from elastic dispersion to nuclear fission; to facilitate the neutrons detection, is necessary to moderate them until reaching electronic equilibrium with the medium at standard conditions of pressure and temperature, in this state the total cross section of the {sup 3}He is large. The objective of the present work was to estimate the response matrix of a proportional detector of {sup 3}He using regular volumes of moderator through Monte Carlo methods. Neutron monoenergetic sources with energies of 10{sup -9} to 20 MeV and polyethylene moderators of different sizes were used. The calculations were made with the MCNP5 code; the number of stories for each detector-moderator combination was large enough to obtain errors less than 1.5%. We found that for small moderators the highest response is obtained for lower energy neutrons, when increasing the moderator dimension we observe that the response decreases for neutrons of lower energy and increases for higher energy neutrons. The total sum of the responses of each moderator allows obtaining a response close to a constant function. (Author)
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. ...
On Improving Convergence Rates for Nonnegative Kernel Density Estimators
Terrell, George R.; Scott, David W.
1980-01-01
To improve the rate of decrease of integrated mean square error for nonparametric kernel density estimators beyond $0(n^{-\\frac{4}{5}}),$ we must relax the constraint that the density estimate be a bonafide density function, that is, be nonnegative and integrate to one. All current methods for kernel (and orthogonal series) estimators relax the nonnegativity constraint. In this paper we show how to achieve similar improvement by relaxing the integral constraint only. This is important in appl...
Graphical analyses of connected-kernel scattering equations
International Nuclear Information System (INIS)
Picklesimer, A.
1982-10-01
Simple graphical techniques are employed to obtain a new (simultaneous) derivation of a large class of connected-kernel scattering equations. This class includes the Rosenberg, Bencze-Redish-Sloan, and connected-kernel multiple scattering equations as well as a host of generalizations of these and other equations. The graphical method also leads to a new, simplified form for some members of the class and elucidates the general structural features of the entire class
Enhanced manifold regularization for semi-supervised classification.
Gan, Haitao; Luo, Zhizeng; Fan, Yingle; Sang, Nong
2016-06-01
Manifold regularization (MR) has become one of the most widely used approaches in the semi-supervised learning field. It has shown superiority by exploiting the local manifold structure of both labeled and unlabeled data. The manifold structure is modeled by constructing a Laplacian graph and then incorporated in learning through a smoothness regularization term. Hence the labels of labeled and unlabeled data vary smoothly along the geodesics on the manifold. However, MR has ignored the discriminative ability of the labeled and unlabeled data. To address the problem, we propose an enhanced MR framework for semi-supervised classification in which the local discriminative information of the labeled and unlabeled data is explicitly exploited. To make full use of labeled data, we firstly employ a semi-supervised clustering method to discover the underlying data space structure of the whole dataset. Then we construct a local discrimination graph to model the discriminative information of labeled and unlabeled data according to the discovered intrinsic structure. Therefore, the data points that may be from different clusters, though similar on the manifold, are enforced far away from each other. Finally, the discrimination graph is incorporated into the MR framework. In particular, we utilize semi-supervised fuzzy c-means and Laplacian regularized Kernel minimum squared error for semi-supervised clustering and classification, respectively. Experimental results on several benchmark datasets and face recognition demonstrate the effectiveness of our proposed method.
Effective Alternating Direction Optimization Methods for Sparsity-Constrained Blind Image Deblurring
Directory of Open Access Journals (Sweden)
Naixue Xiong
2017-01-01
Full Text Available Single-image blind deblurring for imaging sensors in the Internet of Things (IoT is a challenging ill-conditioned inverse problem, which requires regularization techniques to stabilize the image restoration process. The purpose is to recover the underlying blur kernel and latent sharp image from only one blurred image. Under many degraded imaging conditions, the blur kernel could be considered not only spatially sparse, but also piecewise smooth with the support of a continuous curve. By taking advantage of the hybrid sparse properties of the blur kernel, a hybrid regularization method is proposed in this paper to robustly and accurately estimate the blur kernel. The effectiveness of the proposed blur kernel estimation method is enhanced by incorporating both the L 1 -norm of kernel intensity and the squared L 2 -norm of the intensity derivative. Once the accurate estimation of the blur kernel is obtained, the original blind deblurring can be simplified to the direct deconvolution of blurred images. To guarantee robust non-blind deconvolution, a variational image restoration model is presented based on the L 1 -norm data-fidelity term and the total generalized variation (TGV regularizer of second-order. All non-smooth optimization problems related to blur kernel estimation and non-blind deconvolution are effectively handled by using the alternating direction method of multipliers (ADMM-based numerical methods. Comprehensive experiments on both synthetic and realistic datasets have been implemented to compare the proposed method with several state-of-the-art methods. The experimental comparisons have illustrated the satisfactory imaging performance of the proposed method in terms of quantitative and qualitative evaluations.
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...
DEFF Research Database (Denmark)
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....
John A. D. Appleby
2010-01-01
We consider the rate of convergence to equilibrium of Volterra integrodifferential equations with infinite memory. We show that if the kernel of Volterra operator is regularly varying at infinity, and the initial history is regularly varying at minus infinity, then the rate of convergence to the equilibrium is regularly varying at infinity, and the exact pointwise rate of convergence can be determined in terms of the rate of decay of the kernel and the rate of growth of the initial history. ...
Kernel-based tests for joint independence
DEFF Research Database (Denmark)
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...
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.
Adaptive Regularization of Neural Classifiers
DEFF Research Database (Denmark)
Andersen, Lars Nonboe; Larsen, Jan; Hansen, Lars Kai
1997-01-01
We present a regularization scheme which iteratively adapts the regularization parameters by minimizing the validation error. It is suggested to use the adaptive regularization scheme in conjunction with optimal brain damage pruning to optimize the architecture and to avoid overfitting. Furthermo......, we propose an improved neural classification architecture eliminating an inherent redundancy in the widely used SoftMax classification network. Numerical results demonstrate the viability of the method...
Coordinate-invariant regularization
International Nuclear Information System (INIS)
Halpern, M.B.
1987-01-01
A general phase-space framework for coordinate-invariant regularization is given. The development is geometric, with all regularization contained in regularized DeWitt Superstructures on field deformations. Parallel development of invariant coordinate-space regularization is obtained by regularized functional integration of the momenta. As representative examples of the general formulation, the regularized general non-linear sigma model and regularized quantum gravity are discussed. copyright 1987 Academic Press, Inc
Directory of Open Access Journals (Sweden)
Jing Niu
2013-01-01
reproducing kernel on infinite interval is obtained concisely in polynomial form for the first time. Furthermore, as a particular effective application of this method, we give an explicit representation formula for calculation of reproducing kernel in reproducing kernel space with boundary value conditions.
Omnibus risk assessment via accelerated failure time kernel machine modeling.
Sinnott, Jennifer A; Cai, Tianxi
2013-12-01
Integrating genomic information with traditional clinical risk factors to improve the prediction of disease outcomes could profoundly change the practice of medicine. However, the large number of potential markers and possible complexity of the relationship between markers and disease make it difficult to construct accurate risk prediction models. Standard approaches for identifying important markers often rely on marginal associations or linearity assumptions and may not capture non-linear or interactive effects. In recent years, much work has been done to group genes into pathways and networks. Integrating such biological knowledge into statistical learning could potentially improve model interpretability and reliability. One effective approach is to employ a kernel machine (KM) framework, which can capture nonlinear effects if nonlinear kernels are used (Scholkopf and Smola, 2002; Liu et al., 2007, 2008). For survival outcomes, KM regression modeling and testing procedures have been derived under a proportional hazards (PH) assumption (Li and Luan, 2003; Cai, Tonini, and Lin, 2011). In this article, we derive testing and prediction methods for KM regression under the accelerated failure time (AFT) model, a useful alternative to the PH model. We approximate the null distribution of our test statistic using resampling procedures. When multiple kernels are of potential interest, it may be unclear in advance which kernel to use for testing and estimation. We propose a robust Omnibus Test that combines information across kernels, and an approach for selecting the best kernel for estimation. The methods are illustrated with an application in breast cancer. © 2013, The International Biometric Society.
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
Stochastic dynamic modeling of regular and slow earthquakes
Aso, N.; Ando, R.; Ide, S.
2017-12-01
Both regular and slow earthquakes are slip phenomena on plate boundaries and are simulated by a (quasi-)dynamic modeling [Liu and Rice, 2005]. In these numerical simulations, spatial heterogeneity is usually considered not only for explaining real physical properties but also for evaluating the stability of the calculations or the sensitivity of the results on the condition. However, even though we discretize the model space with small grids, heterogeneity at smaller scales than the grid size is not considered in the models with deterministic governing equations. To evaluate the effect of heterogeneity at the smaller scales we need to consider stochastic interactions between slip and stress in a dynamic modeling. Tidal stress is known to trigger or affect both regular and slow earthquakes [Yabe et al., 2015; Ide et al., 2016], and such an external force with fluctuation can also be considered as a stochastic external force. A healing process of faults may also be stochastic, so we introduce stochastic friction law. In the present study, we propose a stochastic dynamic model to explain both regular and slow earthquakes. We solve mode III problem, which corresponds to the rupture propagation along the strike direction. We use BIEM (boundary integral equation method) scheme to simulate slip evolution, but we add stochastic perturbations in the governing equations, which is usually written in a deterministic manner. As the simplest type of perturbations, we adopt Gaussian deviations in the formulation of the slip-stress kernel, external force, and friction. By increasing the amplitude of perturbations of the slip-stress kernel, we reproduce complicated rupture process of regular earthquakes including unilateral and bilateral ruptures. By perturbing external force, we reproduce slow rupture propagation at a scale of km/day. The slow propagation generated by a combination of fast interaction at S-wave velocity is analogous to the kinetic theory of gasses: thermal
Progressive image denoising through hybrid graph Laplacian regularization: a unified framework.
Liu, Xianming; Zhai, Deming; Zhao, Debin; Zhai, Guangtao; Gao, Wen
2014-04-01
Recovering images from corrupted observations is necessary for many real-world applications. In this paper, we propose a unified framework to perform progressive image recovery based on hybrid graph Laplacian regularized regression. We first construct a multiscale representation of the target image by Laplacian pyramid, then progressively recover the degraded image in the scale space from coarse to fine so that the sharp edges and texture can be eventually recovered. On one hand, within each scale, a graph Laplacian regularization model represented by implicit kernel is learned, which simultaneously minimizes the least square error on the measured samples and preserves the geometrical structure of the image data space. In this procedure, the intrinsic manifold structure is explicitly considered using both measured and unmeasured samples, and the nonlocal self-similarity property is utilized as a fruitful resource for abstracting a priori knowledge of the images. On the other hand, between two successive scales, the proposed model is extended to a projected high-dimensional feature space through explicit kernel mapping to describe the interscale correlation, in which the local structure regularity is learned and propagated from coarser to finer scales. In this way, the proposed algorithm gradually recovers more and more image details and edges, which could not been recovered in previous scale. We test our algorithm on one typical image recovery task: impulse noise removal. Experimental results on benchmark test images demonstrate that the proposed method achieves better performance than state-of-the-art algorithms.
Kernel versions of some orthogonal transformations
DEFF Research Database (Denmark)
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
DEFF Research Database (Denmark)
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...
Online co-regularized algorithms
Ruijter, T. de; Tsivtsivadze, E.; Heskes, T.
2012-01-01
We propose an online co-regularized learning algorithm for classification and regression tasks. We demonstrate that by sequentially co-regularizing prediction functions on unlabeled data points, our algorithm provides improved performance in comparison to supervised methods on several UCI benchmarks
Resummed memory kernels in generalized system-bath master equations
International Nuclear Information System (INIS)
Mavros, Michael G.; Van Voorhis, Troy
2014-01-01
Generalized master equations provide a concise formalism for studying reduced population dynamics. Usually, these master equations require a perturbative expansion of the memory kernels governing the dynamics; in order to prevent divergences, these expansions must be resummed. Resummation techniques of perturbation series are ubiquitous in physics, but they have not been readily studied for the time-dependent memory kernels used in generalized master equations. In this paper, we present a comparison of different resummation techniques for such memory kernels up to fourth order. We study specifically the spin-boson Hamiltonian as a model system bath Hamiltonian, treating the diabatic coupling between the two states as a perturbation. A novel derivation of the fourth-order memory kernel for the spin-boson problem is presented; then, the second- and fourth-order kernels are evaluated numerically for a variety of spin-boson parameter regimes. We find that resumming the kernels through fourth order using a Padé approximant results in divergent populations in the strong electronic coupling regime due to a singularity introduced by the nature of the resummation, and thus recommend a non-divergent exponential resummation (the “Landau-Zener resummation” of previous work). The inclusion of fourth-order effects in a Landau-Zener-resummed kernel is shown to improve both the dephasing rate and the obedience of detailed balance over simpler prescriptions like the non-interacting blip approximation, showing a relatively quick convergence on the exact answer. The results suggest that including higher-order contributions to the memory kernel of a generalized master equation and performing an appropriate resummation can provide a numerically-exact solution to system-bath dynamics for a general spectral density, opening the way to a new class of methods for treating system-bath dynamics
MULTITASKER, Multitasking Kernel for C and FORTRAN Under UNIX
International Nuclear Information System (INIS)
Brooks, E.D. III
1988-01-01
1 - Description of program or function: MULTITASKER implements a multitasking kernel for the C and FORTRAN programming languages that runs under UNIX. The kernel provides a multitasking environment which serves two purposes. The first is to provide an efficient portable environment for the development, debugging, and execution of production multiprocessor programs. The second is to provide a means of evaluating the performance of a multitasking program on model multiprocessor hardware. The performance evaluation features require no changes in the application program source and are implemented as a set of compile- and run-time options in the kernel. 2 - Method of solution: The FORTRAN interface to the kernel is identical in function to the CRI multitasking package provided for the Cray XMP. This provides a migration path to high speed (but small N) multiprocessors once the application has been coded and debugged. With use of the UNIX m4 macro preprocessor, source compatibility can be achieved between the UNIX code development system and the target Cray multiprocessor. The kernel also provides a means of evaluating a program's performance on model multiprocessors. Execution traces may be obtained which allow the user to determine kernel overhead, memory conflicts between various tasks, and the average concurrency being exploited. The kernel may also be made to switch tasks every cpu instruction with a random execution ordering. This allows the user to look for unprotected critical regions in the program. These features, implemented as a set of compile- and run-time options, cause extra execution overhead which is not present in the standard production version of the kernel
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.
Multilinear Graph Embedding: Representation and Regularization for Images.
Chen, Yi-Lei; Hsu, Chiou-Ting
2014-02-01
Given a set of images, finding a compact and discriminative representation is still a big challenge especially when multiple latent factors are hidden in the way of data generation. To represent multifactor images, although multilinear models are widely used to parameterize the data, most methods are based on high-order singular value decomposition (HOSVD), which preserves global statistics but interprets local variations inadequately. To this end, we propose a novel method, called multilinear graph embedding (MGE), as well as its kernelization MKGE to leverage the manifold learning techniques into multilinear models. Our method theoretically links the linear, nonlinear, and multilinear dimensionality reduction. We also show that the supervised MGE encodes informative image priors for image regularization, provided that an image is represented as a high-order tensor. From our experiments on face and gait recognition, the superior performance demonstrates that MGE better represents multifactor images than classic methods, including HOSVD and its variants. In addition, the significant improvement in image (or tensor) completion validates the potential of MGE for image regularization.
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.
Integral equations with contrasting kernels
Directory of Open Access Journals (Sweden)
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.
International Nuclear Information System (INIS)
Cho, Bumhee; Cho, Nam Zin
2015-01-01
In this study, the steady-state p-CMFD adjoint flux is used as the weighting function to obtain PK parameters instead of the computationally expensive transport adjoint angular flux. Several numerical problems are investigated to see the capability of the PCQS method applied to the NLG iteration. CRX-2K adopts the nonoverlapping local/global (NLG) iterative method with the 2-D/1-D fusion transport kernel and the global p-CMFD wrapper. The parallelization of the NLG iteration has been recently implemented in CRX-2K and several numerical results are reported in a companion paper. However, the direct time discretization leads to a fine time step size to acquire an accurate transient solution, and the step size involved in the transport transient calculations is millisecond-order. Therefore, the transient calculations need much longer computing time than the steady-state calculation. To increase the time step size, Predictor-Corrector Quasi-Static (PCQS) method can be one option to apply to the NLG iteration. The PCQS method is a linear algorithm, so the shape function does not need to be updated more than once at a specific time step like a conventional quasi-static (QS) family such as Improved Quasi-Static (IQS) method. Moreover, the shape function in the PCQS method directly comes from the direct transport calculation (with a large time step), so one can easily implement the PCQS method in an existing transient transport code. Any QS method needs to solve the amplitude function in the form of the point kinetics (PK) equations, and accurate PK parameters can be obtained by the transport steady-state adjoint angular flux as a weighting function. The PCQS method is applied to the transient NLG iteration with the 2-D/1-D fusion transport kernel and the global p-CMFD wrapper, and has been implemented in CRX-2K. In the numerical problems, the PCQS method with the NLG iteration shows more accurate solutions compared to the direct transient calculations with large time step
Pencil kernel correction and residual error estimation for quality-index-based dose calculations
International Nuclear Information System (INIS)
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
International Nuclear Information System (INIS)
Rijssel, Jos van; Kuipers, Bonny W.M.; Erné, Ben H.
2014-01-01
A numerical inversion method known from the analysis of light scattering by colloidal dispersions is now applied to magnetization curves of ferrofluids. The distribution of magnetic particle sizes or dipole moments is determined without assuming that the distribution is unimodal or of a particular shape. The inversion method enforces positive number densities via a non-negative least squares procedure. It is tested successfully on experimental and simulated data for ferrofluid samples with known multimodal size distributions. The created computer program MINORIM is made available on the web. - Highlights: • A method from light scattering is applied to analyze ferrofluid magnetization curves. • A magnetic size distribution is obtained without prior assumption of its shape. • The method is tested successfully on ferrofluids with a known size distribution. • The practical limits of the method are explored with simulated data including noise. • This method is implemented in the program MINORIM, freely available online
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.
Unsupervised multiple kernel learning for heterogeneous data integration.
Mariette, Jérôme; Villa-Vialaneix, Nathalie
2018-03-15
Recent high-throughput sequencing advances have expanded the breadth of available omics datasets and the integrated analysis of multiple datasets obtained on the same samples has allowed to gain important insights in a wide range of applications. However, the integration of various sources of information remains a challenge for systems biology since produced datasets are often of heterogeneous types, with the need of developing generic methods to take their different specificities into account. We propose a multiple kernel framework that allows to integrate multiple datasets of various types into a single exploratory analysis. Several solutions are provided to learn either a consensus meta-kernel or a meta-kernel that preserves the original topology of the datasets. We applied our framework to analyse two public multi-omics datasets. First, the multiple metagenomic datasets, collected during the TARA Oceans expedition, was explored to demonstrate that our method is able to retrieve previous findings in a single kernel PCA as well as to provide a new image of the sample structures when a larger number of datasets are included in the analysis. To perform this analysis, a generic procedure is also proposed to improve the interpretability of the kernel PCA in regards with the original data. Second, the multi-omics breast cancer datasets, provided by The Cancer Genome Atlas, is analysed using a kernel Self-Organizing Maps with both single and multi-omics strategies. The comparison of these two approaches demonstrates the benefit of our integration method to improve the representation of the studied biological system. Proposed methods are available in the R package mixKernel, released on CRAN. It is fully compatible with the mixOmics package and a tutorial describing the approach can be found on mixOmics web site http://mixomics.org/mixkernel/. jerome.mariette@inra.fr or nathalie.villa-vialaneix@inra.fr. Supplementary data are available at Bioinformatics online.
Model selection for Gaussian kernel PCA denoising
DEFF Research Database (Denmark)
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...
Directory of Open Access Journals (Sweden)
R. MEDEIROS
Full Text Available ABSTRACT This study was conducted with the aim of evaluating the influence of different methods for end surface preparation of compressive strength test specimens. Four different methods were compared: a mechanical wear method through grinding using a diamond wheel established by NBR 5738; a mechanical wear method using a diamond saw which is established by NM 77; an unbonded system using neoprene pads in metal retainer rings established by C1231 and a bonded capping method with sulfur mortar established by NBR 5738 and by NM 77. To develop this research, 4 concrete mixes were determined with different strength levels, 2 of group 1 and 2 of group 2 strength levels established by NBR 8953. Group 1 consists of classes C20 to C50, 5 in 5MPa, also known as normal strength concrete. Group 2 is comprised of class C55, C60 to C100, 10 in 10 MPa, also known as high strength concrete. Compression tests were carried out at 7 and 28 days for the 4 surface preparation methods. The results of this study indicate that the method established by NBR 5738 is the most effective among the 4 strengths considered, once it presents lower dispersion of values obtained from the tests, measured by the coefficient of variation and, in almost all cases, it demonstrates the highest mean of rupture test. The method described by NBR 5738 achieved the expected strength level in all tests.
A survey of kernel-type estimators for copula and their applications
Sumarjaya, I. W.
2017-10-01
Copulas have been widely used to model nonlinear dependence structure. Main applications of copulas include areas such as finance, insurance, hydrology, rainfall to name but a few. The flexibility of copula allows researchers to model dependence structure beyond Gaussian distribution. Basically, a copula is a function that couples multivariate distribution functions to their one-dimensional marginal distribution functions. In general, there are three methods to estimate copula. These are parametric, nonparametric, and semiparametric method. In this article we survey kernel-type estimators for copula such as mirror reflection kernel, beta kernel, transformation method and local likelihood transformation method. Then, we apply these kernel methods to three stock indexes in Asia. The results of our analysis suggest that, albeit variation in information criterion values, the local likelihood transformation method performs better than the other kernel methods.
PERI - auto-tuning memory-intensive kernels for multicore
International Nuclear Information System (INIS)
Williams, S; Carter, J; Oliker, L; Shalf, J; Yelick, K; Bailey, D; Datta, K
2008-01-01
We present an auto-tuning approach to optimize application performance on emerging multicore architectures. The methodology extends the idea of search-based performance optimizations, popular in linear algebra and FFT libraries, to application-specific computational kernels. Our work applies this strategy to sparse matrix vector multiplication (SpMV), the explicit heat equation PDE on a regular grid (Stencil), and a lattice Boltzmann application (LBMHD). We explore one of the broadest sets of multicore architectures in the high-performance computing literature, including the Intel Xeon Clovertown, AMD Opteron Barcelona, Sun Victoria Falls, and the Sony-Toshiba-IBM (STI) Cell. Rather than hand-tuning each kernel for each system, we develop a code generator for each kernel that allows us identify a highly optimized version for each platform, while amortizing the human programming effort. Results show that our auto-tuned kernel applications often achieve a better than 4x improvement compared with the original code. Additionally, we analyze a Roofline performance model for each platform to reveal hardware bottlenecks and software challenges for future multicore systems and applications
PERI - Auto-tuning Memory Intensive Kernels for Multicore
Energy Technology Data Exchange (ETDEWEB)
Bailey, David H; Williams, Samuel; Datta, Kaushik; Carter, Jonathan; Oliker, Leonid; Shalf, John; Yelick, Katherine; Bailey, David H
2008-06-24
We present an auto-tuning approach to optimize application performance on emerging multicore architectures. The methodology extends the idea of search-based performance optimizations, popular in linear algebra and FFT libraries, to application-specific computational kernels. Our work applies this strategy to Sparse Matrix Vector Multiplication (SpMV), the explicit heat equation PDE on a regular grid (Stencil), and a lattice Boltzmann application (LBMHD). We explore one of the broadest sets of multicore architectures in the HPC literature, including the Intel Xeon Clovertown, AMD Opteron Barcelona, Sun Victoria Falls, and the Sony-Toshiba-IBM (STI) Cell. Rather than hand-tuning each kernel for each system, we develop a code generator for each kernel that allows us to identify a highly optimized version for each platform, while amortizing the human programming effort. Results show that our auto-tuned kernel applications often achieve a better than 4X improvement compared with the original code. Additionally, we analyze a Roofline performance model for each platform to reveal hardware bottlenecks and software challenges for future multicore systems and applications.
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
International Nuclear Information System (INIS)
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.
Fixed kernel regression for voltammogram feature extraction
International Nuclear Information System (INIS)
Acevedo Rodriguez, F J; López-Sastre, R J; Gil-Jiménez, P; Maldonado Bascón, S; Ruiz-Reyes, N
2009-01-01
Cyclic voltammetry is an electroanalytical technique for obtaining information about substances under analysis without the need for complex flow systems. However, classifying the information in voltammograms obtained using this technique is difficult. In this paper, we propose the use of fixed kernel regression as a method for extracting features from these voltammograms, reducing the information to a few coefficients. The proposed approach has been applied to a wine classification problem with accuracy rates of over 98%. Although the method is described here for extracting voltammogram information, it can be used for other types of signals
DEFF Research Database (Denmark)
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....
Online learning control using adaptive critic designs with sparse kernel machines.
Xu, Xin; Hou, Zhongsheng; Lian, Chuanqiang; He, Haibo
2013-05-01
In the past decade, adaptive critic designs (ACDs), including heuristic dynamic programming (HDP), dual heuristic programming (DHP), and their action-dependent ones, have been widely studied to realize online learning control of dynamical systems. However, because neural networks with manually designed features are commonly used to deal with continuous state and action spaces, the generalization capability and learning efficiency of previous ACDs still need to be improved. In this paper, a novel framework of ACDs with sparse kernel machines is presented by integrating kernel methods into the critic of ACDs. To improve the generalization capability as well as the computational efficiency of kernel machines, a sparsification method based on the approximately linear dependence analysis is used. Using the sparse kernel machines, two kernel-based ACD algorithms, that is, kernel HDP (KHDP) and kernel DHP (KDHP), are proposed and their performance is analyzed both theoretically and empirically. Because of the representation learning and generalization capability of sparse kernel machines, KHDP and KDHP can obtain much better performance than previous HDP and DHP with manually designed neural networks. Simulation and experimental results of two nonlinear control problems, that is, a continuous-action inverted pendulum problem and a ball and plate control problem, demonstrate the effectiveness of the proposed kernel ACD methods.
Sparse kernel orthonormalized PLS for feature extraction in large datasets
DEFF Research Database (Denmark)
Arenas-García, Jerónimo; Petersen, Kaare Brandt; Hansen, Lars Kai
2006-01-01
In this paper we are presenting a novel multivariate analysis method for large scale problems. Our scheme is based on a novel kernel orthonormalized partial least squares (PLS) variant for feature extraction, imposing sparsity constrains in the solution to improve scalability. The algorithm...... is tested on a benchmark of UCI data sets, and on the analysis of integrated short-time music features for genre prediction. The upshot is that the method has strong expressive power even with rather few features, is clearly outperforming the ordinary kernel PLS, and therefore is an appealing method...
International Nuclear Information System (INIS)
Malykhin, V.M.; Ivanova, N.I.
1981-01-01
It is shown that when assessing the necessary periodicity of internal irradiation monitoring, it is required to take account of the nature (rhythm) of radionuclide intake to the organism during the monitoring period, the effective period of radionuclide biological half-life, its activity in the organism, sensitivity of the technique applied and the labour-consumig character of the monitoring method [ru
Ganzherli, N. M.; Gulyaev, S. N.; Gurin, A. S.; Kramushchenko, D. D.; Maurer, I. A.; Chernykh, D. F.
2009-07-01
The formation of diffusers and microlens rasters on silver halide emulsions by holographic methods is considered. Two techniques for converting amplitude holographic recording to relief-phase recording, selective curing and irradiation of the emulsion gelatin by short-wavelength UV radiation, are compared.
Generalized synthetic kernel approximation for elastic moderation of fast neutrons
International Nuclear Information System (INIS)
Yamamoto, Koji; Sekiya, Tamotsu; Yamamura, Yasunori.
1975-01-01
A method of synthetic kernel approximation is examined in some detail with a view to simplifying the treatment of the elastic moderation of fast neutrons. A sequence of unified kernel (fsub(N)) is introduced, which is then divided into two subsequences (Wsub(n)) and (Gsub(n)) according to whether N is odd (Wsub(n)=fsub(2n-1), n=1,2, ...) or even (Gsub(n)=fsub(2n), n=0,1, ...). The W 1 and G 1 kernels correspond to the usual Wigner and GG kernels, respectively, and the Wsub(n) and Gsub(n) kernels for n>=2 represent generalizations thereof. It is shown that the Wsub(n) kernel solution with a relatively small n (>=2) is superior on the whole to the Gsub(n) kernel solution for the same index n, while both converge to the exact values with increasing n. To evaluate the collision density numerically and rapidly, a simple recurrence formula is derived. In the asymptotic region (except near resonances), this recurrence formula allows calculation with a relatively coarse mesh width whenever hsub(a)<=0.05 at least. For calculations in the transient lethargy region, a mesh width of order epsilon/10 is small enough to evaluate the approximate collision density psisub(N) with an accuracy comparable to that obtained analytically. It is shown that, with the present method, an order of approximation of about n=7 should yield a practically correct solution diviating not more than 1% in collision density. (auth.)
Study of the convergence behavior of the complex kernel least mean square algorithm.
Paul, Thomas K; Ogunfunmi, Tokunbo
2013-09-01
The complex kernel least mean square (CKLMS) algorithm is recently derived and allows for online kernel adaptive learning for complex data. Kernel adaptive methods can be used in finding solutions for neural network and machine learning applications. The derivation of CKLMS involved the development of a modified Wirtinger calculus for Hilbert spaces to obtain the cost function gradient. We analyze the convergence of the CKLMS with different kernel forms for complex data. The expressions obtained enable us to generate theory-predicted mean-square error curves considering the circularity of the complex input signals and their effect on nonlinear learning. Simulations are used for verifying the analysis results.
Regularizing portfolio optimization
International Nuclear Information System (INIS)
Still, Susanne; Kondor, Imre
2010-01-01
The optimization of large portfolios displays an inherent instability due to estimation error. This poses a fundamental problem, because solutions that are not stable under sample fluctuations may look optimal for a given sample, but are, in effect, very far from optimal with respect to the average risk. In this paper, we approach the problem from the point of view of statistical learning theory. The occurrence of the instability is intimately related to over-fitting, which can be avoided using known regularization methods. We show how regularized portfolio optimization with the expected shortfall as a risk measure is related to support vector regression. The budget constraint dictates a modification. We present the resulting optimization problem and discuss the solution. The L2 norm of the weight vector is used as a regularizer, which corresponds to a diversification 'pressure'. This means that diversification, besides counteracting downward fluctuations in some assets by upward fluctuations in others, is also crucial because it improves the stability of the solution. The approach we provide here allows for the simultaneous treatment of optimization and diversification in one framework that enables the investor to trade off between the two, depending on the size of the available dataset.
Regularizing portfolio optimization
Still, Susanne; Kondor, Imre
2010-07-01
The optimization of large portfolios displays an inherent instability due to estimation error. This poses a fundamental problem, because solutions that are not stable under sample fluctuations may look optimal for a given sample, but are, in effect, very far from optimal with respect to the average risk. In this paper, we approach the problem from the point of view of statistical learning theory. The occurrence of the instability is intimately related to over-fitting, which can be avoided using known regularization methods. We show how regularized portfolio optimization with the expected shortfall as a risk measure is related to support vector regression. The budget constraint dictates a modification. We present the resulting optimization problem and discuss the solution. The L2 norm of the weight vector is used as a regularizer, which corresponds to a diversification 'pressure'. This means that diversification, besides counteracting downward fluctuations in some assets by upward fluctuations in others, is also crucial because it improves the stability of the solution. The approach we provide here allows for the simultaneous treatment of optimization and diversification in one framework that enables the investor to trade off between the two, depending on the size of the available dataset.
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.
Improving prediction of heterodimeric protein complexes using combination with pairwise kernel.
Ruan, Peiying; Hayashida, Morihiro; Akutsu, Tatsuya; Vert, Jean-Philippe
2018-02-19
Since many proteins become functional only after they interact with their partner proteins and form protein complexes, it is essential to identify the sets of proteins that form complexes. Therefore, several computational methods have been proposed to predict complexes from the topology and structure of experimental protein-protein interaction (PPI) network. These methods work well to predict complexes involving at least three proteins, but generally fail at identifying complexes involving only two different proteins, called heterodimeric complexes or heterodimers. There is however an urgent need for efficient methods to predict heterodimers, since the majority of known protein complexes are precisely heterodimers. In this paper, we use three promising kernel functions, Min kernel and two pairwise kernels, which are Metric Learning Pairwise Kernel (MLPK) and Tensor Product Pairwise Kernel (TPPK). We also consider the normalization forms of Min kernel. Then, we combine Min kernel or its normalization form and one of the pairwise kernels by plugging. We applied kernels based on PPI, domain, phylogenetic profile, and subcellular localization properties to predicting heterodimers. Then, we evaluate our method by employing C-Support Vector Classification (C-SVC), carrying out 10-fold cross-validation, and calculating the average F-measures. The results suggest that the combination of normalized-Min-kernel and MLPK leads to the best F-measure and improved the performance of our previous work, which had been the best existing method so far. We propose new methods to predict heterodimers, using a machine learning-based approach. We train a support vector machine (SVM) to discriminate interacting vs non-interacting protein pairs, based on informations extracted from PPI, domain, phylogenetic profiles and subcellular localization. We evaluate in detail new kernel functions to encode these data, and report prediction performance that outperforms the state-of-the-art.
Weighted Feature Gaussian Kernel SVM for Emotion Recognition.
Wei, Wei; Jia, Qingxuan
2016-01-01
Emotion recognition with weighted feature based on facial expression is a challenging research topic and has attracted great attention in the past few years. This paper presents a novel method, utilizing subregion recognition rate to weight kernel function. First, we divide the facial expression image into some uniform subregions and calculate corresponding recognition rate and weight. Then, we get a weighted feature Gaussian kernel function and construct a classifier based on Support Vector Machine (SVM). At last, the experimental results suggest that the approach based on weighted feature Gaussian kernel function has good performance on the correct rate in emotion recognition. The experiments on the extended Cohn-Kanade (CK+) dataset show that our method has achieved encouraging recognition results compared to the state-of-the-art methods.
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
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.
The Kernel Estimation in Biosystems Engineering
Directory of Open Access Journals (Sweden)
Esperanza Ayuga Téllez
2008-04-01
Full Text Available In many fields of biosystems engineering, it is common to find works in which statistical information is analysed that violates the basic hypotheses necessary for the conventional forecasting methods. For those situations, it is necessary to find alternative methods that allow the statistical analysis considering those infringements. Non-parametric function estimation includes methods that fit a target function locally, using data from a small neighbourhood of the point. Weak assumptions, such as continuity and differentiability of the target function, are rather used than "a priori" assumption of the global target function shape (e.g., linear or quadratic. In this paper a few basic rules of decision are enunciated, for the application of the non-parametric estimation method. These statistical rules set up the first step to build an interface usermethod for the consistent application of kernel estimation for not expert users. To reach this aim, univariate and multivariate estimation methods and density function were analysed, as well as regression estimators. In some cases the models to be applied in different situations, based on simulations, were defined. Different biosystems engineering applications of the kernel estimation are also analysed in this review.
Complex use of cottonseed kernels
Energy Technology Data Exchange (ETDEWEB)
Glushenkova, A I
1977-01-01
A review with 41 references is made on the manufacture of oil, protein, and other products from cottonseed, the effects of gossypol on protein yield and quality and technology of gossypol removal. A process eliminating thermal treatment of the kernels and permitting the production of oil, proteins, phytin, gossypol, sugar, sterols, phosphatides, tocopherols, and residual shells and baggase is described.
Kernel regression with functional response
Ferraty, Frédéric; Laksaci, Ali; Tadj, Amel; Vieu, Philippe
2011-01-01
We consider kernel regression estimate when both the response variable and the explanatory one are functional. The rates of uniform almost complete convergence are stated as function of the small ball probability of the predictor and as function of the entropy of the set on which uniformity is obtained.
GRIM : Leveraging GPUs for Kernel integrity monitoring
Koromilas, Lazaros; Vasiliadis, Giorgos; Athanasopoulos, Ilias; Ioannidis, Sotiris
2016-01-01
Kernel rootkits can exploit an operating system and enable future accessibility and control, despite all recent advances in software protection. A promising defense mechanism against rootkits is Kernel Integrity Monitor (KIM) systems, which inspect the kernel text and data to discover any malicious
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
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
von Larcher, Thomas; Blome, Therese; Klein, Rupert; Schneider, Reinhold; Wolf, Sebastian; Huber, Benjamin
2016-04-01
Handling high-dimensional data sets like they occur e.g. in turbulent flows or in multiscale behaviour of certain types in Geosciences are one of the big challenges in numerical analysis and scientific computing. A suitable solution is to represent those large data sets in an appropriate compact form. In this context, tensor product decomposition methods currently emerge as an important tool. One reason is that these methods often enable one to attack high-dimensional problems successfully, another that they allow for very compact representations of large data sets. We follow the novel Tensor-Train (TT) decomposition method to support the development of improved understanding of the multiscale behavior and the development of compact storage schemes for solutions of such problems. One long-term goal of the project is the construction of a self-consistent closure for Large Eddy Simulations (LES) of turbulent flows that explicitly exploits the tensor product approach's capability of capturing self-similar structures. Secondly, we focus on a mixed deterministic-stochastic subgrid scale modelling strategy currently under development for application in Finite Volume Large Eddy Simulation (LES) codes. Advanced methods of time series analysis for the databased construction of stochastic models with inherently non-stationary statistical properties and concepts of information theory based on a modified Akaike information criterion and on the Bayesian information criterion for the model discrimination are used to construct surrogate models for the non-resolved flux fluctuations. Vector-valued auto-regressive models with external influences form the basis for the modelling approach [1], [2], [4]. Here, we present the reconstruction capabilities of the two modeling approaches tested against 3D turbulent channel flow data computed by direct numerical simulation (DNS) for an incompressible, isothermal fluid at Reynolds number Reτ = 590 (computed by [3]). References [1] I
Directory of Open Access Journals (Sweden)
Renata Bujak
2016-07-01
Full Text Available Non-targeted metabolomics constitutes a part of systems biology and aims to determine many metabolites in complex biological samples. Datasets obtained in non-targeted metabolomics studies are multivariate and high-dimensional due to the sensitivity of mass spectrometry-based detection methods as well as complexity of biological matrices. Proper selection of variables which contribute into group classification is a crucial step, especially in metabolomics studies which are focused on searching for disease biomarker candidates. In the present study, three different statistical approaches were tested using two metabolomics datasets (RH and PH study. Orthogonal projections to latent structures-discriminant analysis (OPLS-DA without and with multiple testing correction as well as least absolute shrinkage and selection operator (LASSO were tested and compared. For the RH study, OPLS-DA model built without multiple testing correction, selected 46 and 218 variables based on VIP criteria using Pareto and UV scaling, respectively. In the case of the PH study, 217 and 320 variables were selected based on VIP criteria using Pareto and UV scaling, respectively. In the RH study, OPLS-DA model built with multiple testing correction, selected 4 and 19 variables as statistically significant in terms of Pareto and UV scaling, respectively. For PH study, 14 and 18 variables were selected based on VIP criteria in terms of Pareto and UV scaling, respectively. Additionally, the concept and fundaments of the least absolute shrinkage and selection operator (LASSO with bootstrap procedure evaluating reproducibility of results, was demonstrated. In the RH and PH study, the LASSO selected 14 and 4 variables with reproducibility between 99.3% and 100%. However, apart from the popularity of PLS-DA and OPLS-DA methods in metabolomics, it should be highlighted that they do not control type I or type II error, but only arbitrarily establish a cut-off value for PLS-DA loadings
Delimiting areas of endemism through kernel interpolation.
Oliveira, Ubirajara; Brescovit, Antonio D; Santos, Adalberto J
2015-01-01
We propose a new approach for identification of areas of endemism, the Geographical Interpolation of Endemism (GIE), based on kernel spatial interpolation. This method differs from others in being independent of grid cells. This new approach is based on estimating the overlap between the distribution of species through a kernel interpolation of centroids of species distribution and areas of influence defined from the distance between the centroid and the farthest point of occurrence of each species. We used this method to delimit areas of endemism of spiders from Brazil. To assess the effectiveness of GIE, we analyzed the same data using Parsimony Analysis of Endemism and NDM and compared the areas identified through each method. The analyses using GIE identified 101 areas of endemism of spiders in Brazil GIE demonstrated to be effective in identifying areas of endemism in multiple scales, with fuzzy edges and supported by more synendemic species than in the other methods. The areas of endemism identified with GIE were generally congruent with those identified for other taxonomic groups, suggesting that common processes can be responsible for the origin and maintenance of these biogeographic units.
Delimiting areas of endemism through kernel interpolation.
Directory of Open Access Journals (Sweden)
Ubirajara Oliveira
Full Text Available We propose a new approach for identification of areas of endemism, the Geographical Interpolation of Endemism (GIE, based on kernel spatial interpolation. This method differs from others in being independent of grid cells. This new approach is based on estimating the overlap between the distribution of species through a kernel interpolation of centroids of species distribution and areas of influence defined from the distance between the centroid and the farthest point of occurrence of each species. We used this method to delimit areas of endemism of spiders from Brazil. To assess the effectiveness of GIE, we analyzed the same data using Parsimony Analysis of Endemism and NDM and compared the areas identified through each method. The analyses using GIE identified 101 areas of endemism of spiders in Brazil GIE demonstrated to be effective in identifying areas of endemism in multiple scales, with fuzzy edges and supported by more synendemic species than in the other methods. The areas of endemism identified with GIE were generally congruent with those identified for other taxonomic groups, suggesting that common processes can be responsible for the origin and maintenance of these biogeographic units.
Kernel and divergence techniques in high energy physics separations
Bouř, Petr; Kůs, Václav; Franc, Jiří
2017-10-01
Binary decision trees under the Bayesian decision technique are used for supervised classification of high-dimensional data. We present a great potential of adaptive kernel density estimation as the nested separation method of the supervised binary divergence decision tree. Also, we provide a proof of alternative computing approach for kernel estimates utilizing Fourier transform. Further, we apply our method to Monte Carlo data set from the particle accelerator Tevatron at DØ experiment in Fermilab and provide final top-antitop signal separation results. We have achieved up to 82 % AUC while using the restricted feature selection entering the signal separation procedure.
Directory of Open Access Journals (Sweden)
M. Madheswaran
2012-06-01
Full Text Available Modern fighter aircrafts, ships, missiles etc need to be very low Radar Cross Section (RCS designs, to avoid detection by hostile radars. Hence accurate prediction of RCS of complex objects like aircrafts is essential to meet this requirement. A simple and efficient numerical procedure for treating problems of wide band RCS prediction Perfect Electric Conductor (PEC objects is developed using Method of Moment (MoM. Implementation of MoM for prediction of RCS involves solving Electric Field Integral Equation (EFIE for electric current using the vector and scalar potential solutions, which satisfy the boundary condition that the tangential electric field at the boundary of the PEC body is zero. For numerical purposes, the objects are modeled using planar triangular surfaces patches. Set of special sub-domain type basis functions are defined on pairs of adjacent triangular patches. These basis functions yield a current representation free of line or point charges at sub-domain boundaries. Once the current distribution is obtained, dipole model is used to find Scattering field in free space. RCS can be calculated from the scattered and incident fields. Numerical results for a square plate, a cube, and a sphere are presented over a bandwidth.
Evaluating Equating Results: Percent Relative Error for Chained Kernel Equating
Jiang, Yanlin; von Davier, Alina A.; Chen, Haiwen
2012-01-01
This article presents a method for evaluating equating results. Within the kernel equating framework, the percent relative error (PRE) for chained equipercentile equating was computed under the nonequivalent groups with anchor test (NEAT) design. The method was applied to two data sets to obtain the PRE, which can be used to measure equating…
Zhang, Wencan; Leong, Siew Mun; Zhao, Feifei; Zhao, Fangju; Yang, Tiankui; Liu, Shaoquan
2018-05-01
With an interest to enhance the aroma of palm kernel oil (PKO), Viscozyme L, an enzyme complex containing a wide range of carbohydrases, was applied to alter the carbohydrates in palm kernels (PK) to modulate the formation of volatiles upon kernel roasting. After Viscozyme treatment, the content of simple sugars and free amino acids in PK increased by 4.4-fold and 4.5-fold, respectively. After kernel roasting and oil extraction, significantly more 2,5-dimethylfuran, 2-[(methylthio)methyl]-furan, 1-(2-furanyl)-ethanone, 1-(2-furyl)-2-propanone, 5-methyl-2-furancarboxaldehyde and 2-acetyl-5-methylfuran but less 2-furanmethanol and 2-furanmethanol acetate were found in treated PKO; the correlation between their formation and simple sugar profile was estimated by using partial least square regression (PLS1). Obvious differences in pyrroles and Strecker aldehydes were also found between the control and treated PKOs. Principal component analysis (PCA) clearly discriminated the treated PKOs from that of control PKOs on the basis of all volatile compounds. Such changes in volatiles translated into distinct sensory attributes, whereby treated PKO was more caramelic and burnt after aqueous extraction and more nutty, roasty, caramelic and smoky after solvent extraction. Copyright © 2018 Elsevier Ltd. All rights reserved.
Wigner functions defined with Laplace transform kernels.
Oh, Se Baek; Petruccelli, Jonathan C; Tian, Lei; Barbastathis, George
2011-10-24
We propose a new Wigner-type phase-space function using Laplace transform kernels--Laplace kernel Wigner function. Whereas momentum variables are real in the traditional Wigner function, the Laplace kernel Wigner function may have complex momentum variables. Due to the property of the Laplace transform, a broader range of signals can be represented in complex phase-space. We show that the Laplace kernel Wigner function exhibits similar properties in the marginals as the traditional Wigner function. As an example, we use the Laplace kernel Wigner function to analyze evanescent waves supported by surface plasmon polariton. © 2011 Optical Society of America
Review of Palm Kernel Oil Processing And Storage Techniques In South East Nigeria
Directory of Open Access Journals (Sweden)
Okeke CG
2017-06-01
Full Text Available An assessment of palm kernel processing and storage in South-Eastern Nigeria was carried out by investigative survey approach. The survey basically ascertained the extent of mechanization applicable in the area to enable the palm kernel processors and agricultural policy makers, device the modalities for improving palm kernel processing in the area. According to the results obtained from the study, in Abia state, 85% of the respondents use mechanical method while 15% use manual method in cracking their kernels. In Imo state, 83% of the processors use mechanical method while 17% use manual method. In Enugu and Ebonyi state, 70% and 50% of the processors respectively use mechanical method. It is only in Anambra state that greater number of the processors (50% use manual method while 45% use mechanical means. It is observable from the results that palm kernel oil extraction has not received much attention in mechanization. The ANOVA of the palm kernel oil extraction technique in South- East Nigeria showed significant difference in both the study area and oil extraction techniques at 5% level of probability. Results further revealed that in Abia State, 70% of the processors use complete fractional process in refining the palm kernel oil; 25% and 5% respectively use incomplete fractional process and zero refining process. In Anambra, 60% of the processors use complete fractional process and 40% use incomplete fractional process. Zero refining method is not applicable in Anambra state. In Enugu sate, 53% use complete fractional process while 25% and 22% respectively use zero refining and incomplete fractional process in refining the palm kernel oil. Imo state, mostly use complete fractional process (85% in refining palm kernel oil. About 10% use zero refining method while 5% of the processors use incomplete fractional process. Plastic containers and metal drums are dominantly used in most areas in south-east Nigeria for the storage of palm kernel oil.
Ensemble-based forecasting at Horns Rev: Ensemble conversion and kernel dressing
DEFF Research Database (Denmark)
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...
Regularized Biot-Savart Laws for Modeling Magnetic Flux Ropes
Titov, Viacheslav; Downs, Cooper; Mikic, Zoran; Torok, Tibor; Linker, Jon A.
2017-08-01
Many existing models assume that magnetic flux ropes play a key role in solar flares and coronal mass ejections (CMEs). It is therefore important to develop efficient methods for constructing flux-rope configurations constrained by observed magnetic data and the initial morphology of CMEs. As our new step in this direction, we have derived and implemented a compact analytical form that represents the magnetic field of a thin flux rope with an axis of arbitrary shape and a circular cross-section. This form implies that the flux rope carries axial current I and axial flux F, so that the respective magnetic field is a curl of the sum of toroidal and poloidal vector potentials proportional to I and F, respectively. The vector potentials are expressed in terms of Biot-Savart laws whose kernels are regularized at the rope axis. We regularized them in such a way that for a straight-line axis the form provides a cylindrical force-free flux rope with a parabolic profile of the axial current density. So far, we set the shape of the rope axis by tracking the polarity inversion lines of observed magnetograms and estimating its height and other parameters of the rope from a calculated potential field above these lines. In spite of this heuristic approach, we were able to successfully construct pre-eruption configurations for the 2009 February13 and 2011 October 1 CME events. These applications demonstrate that our regularized Biot-Savart laws are indeed a very flexible and efficient method for energizing initial configurations in MHD simulations of CMEs. We discuss possible ways of optimizing the axis paths and other extensions of the method in order to make it more useful and robust.Research supported by NSF, NASA's HSR and LWS Programs, and AFOSR.
Testing Infrastructure for Operating System Kernel Development
DEFF Research Database (Denmark)
Walter, Maxwell; Karlsson, Sven
2014-01-01
Testing is an important part of system development, and to test effectively we require knowledge of the internal state of the system under test. Testing an operating system kernel is a challenge as it is the operating system that typically provides access to this internal state information. Multi......-core kernels pose an even greater challenge due to concurrency and their shared kernel state. In this paper, we present a testing framework that addresses these challenges by running the operating system in a virtual machine, and using virtual machine introspection to both communicate with the kernel...... and obtain information about the system. We have also developed an in-kernel testing API that we can use to develop a suite of unit tests in the kernel. We are using our framework for for the development of our own multi-core research kernel....
Mixed kernel function support vector regression for global sensitivity analysis
Cheng, Kai; Lu, Zhenzhou; Wei, Yuhao; Shi, Yan; Zhou, Yicheng
2017-11-01
Global sensitivity analysis (GSA) plays an important role in exploring the respective effects of input variables on an assigned output response. Amongst the wide sensitivity analyses in literature, the Sobol indices have attracted much attention since they can provide accurate information for most models. In this paper, a mixed kernel function (MKF) based support vector regression (SVR) model is employed to evaluate the Sobol indices at low computational cost. By the proposed derivation, the estimation of the Sobol indices can be obtained by post-processing the coefficients of the SVR meta-model. The MKF is constituted by the orthogonal polynomials kernel function and Gaussian radial basis kernel function, thus the MKF possesses both the global characteristic advantage of the polynomials kernel function and the local characteristic advantage of the Gaussian radial basis kernel function. The proposed approach is suitable for high-dimensional and non-linear problems. Performance of the proposed approach is validated by various analytical functions and compared with the popular polynomial chaos expansion (PCE). Results demonstrate that the proposed approach is an efficient method for global sensitivity analysis.
Semi-supervised learning for ordinal Kernel Discriminant Analysis.
Pérez-Ortiz, M; Gutiérrez, P A; Carbonero-Ruz, M; Hervás-Martínez, C
2016-12-01
Ordinal classification considers those classification problems where the labels of the variable to predict follow a given order. Naturally, labelled data is scarce or difficult to obtain in this type of problems because, in many cases, ordinal labels are given by a user or expert (e.g. in recommendation systems). Firstly, this paper develops a new strategy for ordinal classification where both labelled and unlabelled data are used in the model construction step (a scheme which is referred to as semi-supervised learning). More specifically, the ordinal version of kernel discriminant learning is extended for this setting considering the neighbourhood information of unlabelled data, which is proposed to be computed in the feature space induced by the kernel function. Secondly, a new method for semi-supervised kernel learning is devised in the context of ordinal classification, which is combined with our developed classification strategy to optimise the kernel parameters. The experiments conducted compare 6 different approaches for semi-supervised learning in the context of ordinal classification in a battery of 30 datasets, showing (1) the good synergy of the ordinal version of discriminant analysis and the use of unlabelled data and (2) the advantage of computing distances in the feature space induced by the kernel function. Copyright © 2016 Elsevier Ltd. All rights reserved.
The global kernel k-means algorithm for clustering in feature space.
Tzortzis, Grigorios F; Likas, Aristidis C
2009-07-01
Kernel k-means is an extension of the standard k -means clustering algorithm that identifies nonlinearly separable clusters. In order to overcome the cluster initialization problem associated with this method, we propose the global kernel k-means algorithm, a deterministic and incremental approach to kernel-based clustering. Our method adds one cluster at each stage, through a global search procedure consisting of several executions of kernel k-means from suitable initializations. This algorithm does not depend on cluster initialization, identifies nonlinearly separable clusters, and, due to its incremental nature and search procedure, locates near-optimal solutions avoiding poor local minima. Furthermore, two modifications are developed to reduce the computational cost that do not significantly affect the solution quality. The proposed methods are extended to handle weighted data points, which enables their application to graph partitioning. We experiment with several data sets and the proposed approach compares favorably to kernel k -means with random restarts.
Manifold Regularized Correlation Object Tracking
Hu, Hongwei; Ma, Bo; Shen, Jianbing; Shao, Ling
2017-01-01
In this paper, we propose a manifold regularized correlation tracking method with augmented samples. To make better use of the unlabeled data and the manifold structure of the sample space, a manifold regularization-based correlation filter is introduced, which aims to assign similar labels to neighbor samples. Meanwhile, the regression model is learned by exploiting the block-circulant structure of matrices resulting from the augmented translated samples over multiple base samples cropped fr...
Directory of Open Access Journals (Sweden)
Seyed Bahram Beheshti-Aval
2015-06-01
Full Text Available Displacement Coefficient Method (DCM stipulated in the ASCE 41-06 standard is becoming the preferred method for seismic rehabilitation of buildings in many high-seismic-hazard countries. Applications of the method for non-building constructions such as bridges are beyond the scope of this standard. Thus its application to this kind of structure should be approached with care. Target displacement has reasonable accuracy for buildings with strong columns and weak beams, where there is the development of plastic hinges. Due to high stiffness and strength of the deck relative to the piers in most bridges, this mechanism does not occur, and it is necessary to evaluate the accuracy of DCM for such structures. In this research, an attempt is made to evaluate the credibility of DCM in the ASCE/SEI 41-06 standard for estimating target drifts in concrete regular bridges under strong ground motions. To apply the extension of the method to bridge structures, the definition of new correction factor CB, which should be multiplied to previous coefficients, is required. This novel coefficient can improve the accuracy of the mentioned method in accessing seismic displacement demands. The coefficient is presented for soil types A to D based on NEHRP soil classification. The validity of the modified DCM is examined for several bridges with use of nonlinear dynamic analysis. Good correlation is found between both procedures.
Free energy on a cycle graph and trigonometric deformation of heat kernel traces on odd spheres
Kan, Nahomi; Shiraishi, Kiyoshi
2018-01-01
We consider a possible ‘deformation’ of the trace of the heat kernel on odd dimensional spheres, motivated by the calculation of the free energy of a scalar field on a discretized circle. By using an expansion in terms of the modified Bessel functions, we obtain the values of the free energies after a suitable regularization.
Searching remote homology with spectral clustering with symmetry in neighborhood cluster kernels.
Directory of Open Access Journals (Sweden)
Ujjwal Maulik
Full Text Available Remote homology detection among proteins utilizing only the unlabelled sequences is a central problem in comparative genomics. The existing cluster kernel methods based on neighborhoods and profiles and the Markov clustering algorithms are currently the most popular methods for protein family recognition. The deviation from random walks with inflation or dependency on hard threshold in similarity measure in those methods requires an enhancement for homology detection among multi-domain proteins. We propose to combine spectral clustering with neighborhood kernels in Markov similarity for enhancing sensitivity in detecting homology independent of "recent" paralogs. The spectral clustering approach with new combined local alignment kernels more effectively exploits the unsupervised protein sequences globally reducing inter-cluster walks. When combined with the corrections based on modified symmetry based proximity norm deemphasizing outliers, the technique proposed in this article outperforms other state-of-the-art cluster kernels among all twelve implemented kernels. The comparison with the state-of-the-art string and mismatch kernels also show the superior performance scores provided by the proposed kernels. Similar performance improvement also is found over an existing large dataset. Therefore the proposed spectral clustering framework over combined local alignment kernels with modified symmetry based correction achieves superior performance for unsupervised remote homolog detection even in multi-domain and promiscuous domain proteins from Genolevures database families with better biological relevance. Source code available upon request.sarkar@labri.fr.
Theoretical developments for interpreting kernel spectral clustering from alternative viewpoints
Directory of Open Access Journals (Sweden)
Diego Peluffo-Ordóñez
2017-08-01
Full Text Available To perform an exploration process over complex structured data within unsupervised settings, the so-called kernel spectral clustering (KSC is one of the most recommended and appealing approaches, given its versatility and elegant formulation. In this work, we explore the relationship between (KSC and other well-known approaches, namely normalized cut clustering and kernel k-means. To do so, we first deduce a generic KSC model from a primal-dual formulation based on least-squares support-vector machines (LS-SVM. For experiments, KSC as well as other consider methods are assessed on image segmentation tasks to prove their usability.
Graphical analyses of connected-kernel scattering equations
International Nuclear Information System (INIS)
Picklesimer, A.
1983-01-01
Simple graphical techniques are employed to obtain a new (simultaneous) derivation of a large class of connected-kernel scattering equations. This class includes the Rosenberg, Bencze-Redish-Sloan, and connected-kernel multiple scattering equations as well as a host of generalizations of these and other equations. The basic result is the application of graphical methods to the derivation of interaction-set equations. This yields a new, simplified form for some members of the class and elucidates the general structural features of the entire class
van Dam, Edwin R.; Koolen, Jack H.; Tanaka, Hajime
2016-01-01
This is a survey of distance-regular graphs. We present an introduction to distance-regular graphs for the reader who is unfamiliar with the subject, and then give an overview of some developments in the area of distance-regular graphs since the monograph 'BCN'[Brouwer, A.E., Cohen, A.M., Neumaier,
Nijholt, Antinus
1980-01-01
Culik II and Cogen introduced the class of LR-regular grammars, an extension of the LR(k) grammars. In this paper we consider an analogous extension of the LL(k) grammars called the LL-regular grammars. The relation of this class of grammars to other classes of grammars will be shown. Any LL-regular
Validation of Born Traveltime Kernels
Baig, A. M.; Dahlen, F. A.; Hung, S.
2001-12-01
Most inversions for Earth structure using seismic traveltimes rely on linear ray theory to translate observed traveltime anomalies into seismic velocity anomalies distributed throughout the mantle. However, ray theory is not an appropriate tool to use when velocity anomalies have scale lengths less than the width of the Fresnel zone. In the presence of these structures, we need to turn to a scattering theory in order to adequately describe all of the features observed in the waveform. By coupling the Born approximation to ray theory, the first order dependence of heterogeneity on the cross-correlated traveltimes (described by the Fréchet derivative or, more colourfully, the banana-doughnut kernel) may be determined. To determine for what range of parameters these banana-doughnut kernels outperform linear ray theory, we generate several random media specified by their statistical properties, namely the RMS slowness perturbation and the scale length of the heterogeneity. Acoustic waves are numerically generated from a point source using a 3-D pseudo-spectral wave propagation code. These waves are then recorded at a variety of propagation distances from the source introducing a third parameter to the problem: the number of wavelengths traversed by the wave. When all of the heterogeneity has scale lengths larger than the width of the Fresnel zone, ray theory does as good a job at predicting the cross-correlated traveltime as the banana-doughnut kernels do. Below this limit, wavefront healing becomes a significant effect and ray theory ceases to be effective even though the kernels remain relatively accurate provided the heterogeneity is weak. The study of wave propagation in random media is of a more general interest and we will also show our measurements of the velocity shift and the variance of traveltime compare to various theoretical predictions in a given regime.