Forecasting short-term data center network traffic load with convolutional neural networks

Science.gov (United States)

Ordozgoiti, Bruno; Gómez-Canaval, Sandra

2018-01-01

Efficient resource management in data centers is of central importance to content service providers as 90 percent of the network traffic is expected to go through them in the coming years. In this context we propose the use of convolutional neural networks (CNNs) to forecast short-term changes in the amount of traffic crossing a data center network. This value is an indicator of virtual machine activity and can be utilized to shape the data center infrastructure accordingly. The behaviour of network traffic at the seconds scale is highly chaotic and therefore traditional time-series-analysis approaches such as ARIMA fail to obtain accurate forecasts. We show that our convolutional neural network approach can exploit the non-linear regularities of network traffic, providing significant improvements with respect to the mean absolute and standard deviation of the data, and outperforming ARIMA by an increasingly significant margin as the forecasting granularity is above the 16-second resolution. In order to increase the accuracy of the forecasting model, we exploit the architecture of the CNNs using multiresolution input distributed among separate channels of the first convolutional layer. We validate our approach with an extensive set of experiments using a data set collected at the core network of an Internet Service Provider over a period of 5 months, totalling 70 days of traffic at the one-second resolution. PMID:29408936

Forecasting short-term data center network traffic load with convolutional neural networks.

Science.gov (United States)

Mozo, Alberto; Ordozgoiti, Bruno; Gómez-Canaval, Sandra

2018-01-01

Efficient resource management in data centers is of central importance to content service providers as 90 percent of the network traffic is expected to go through them in the coming years. In this context we propose the use of convolutional neural networks (CNNs) to forecast short-term changes in the amount of traffic crossing a data center network. This value is an indicator of virtual machine activity and can be utilized to shape the data center infrastructure accordingly. The behaviour of network traffic at the seconds scale is highly chaotic and therefore traditional time-series-analysis approaches such as ARIMA fail to obtain accurate forecasts. We show that our convolutional neural network approach can exploit the non-linear regularities of network traffic, providing significant improvements with respect to the mean absolute and standard deviation of the data, and outperforming ARIMA by an increasingly significant margin as the forecasting granularity is above the 16-second resolution. In order to increase the accuracy of the forecasting model, we exploit the architecture of the CNNs using multiresolution input distributed among separate channels of the first convolutional layer. We validate our approach with an extensive set of experiments using a data set collected at the core network of an Internet Service Provider over a period of 5 months, totalling 70 days of traffic at the one-second resolution.

Fully automatic, multiorgan segmentation in normal whole body magnetic resonance imaging (MRI), using classification forests (CFs), convolutional neural networks (CNNs), and a multi-atlas (MA) approach.

Science.gov (United States)

Lavdas, Ioannis; Glocker, Ben; Kamnitsas, Konstantinos; Rueckert, Daniel; Mair, Henrietta; Sandhu, Amandeep; Taylor, Stuart A; Aboagye, Eric O; Rockall, Andrea G

2017-10-01

As part of a program to implement automatic lesion detection methods for whole body magnetic resonance imaging (MRI) in oncology, we have developed, evaluated, and compared three algorithms for fully automatic, multiorgan segmentation in healthy volunteers. The first algorithm is based on classification forests (CFs), the second is based on 3D convolutional neural networks (CNNs) and the third algorithm is based on a multi-atlas (MA) approach. We examined data from 51 healthy volunteers, scanned prospectively with a standardized, multiparametric whole body MRI protocol at 1.5 T. The study was approved by the local ethics committee and written consent was obtained from the participants. MRI data were used as input data to the algorithms, while training was based on manual annotation of the anatomies of interest by clinical MRI experts. Fivefold cross-validation experiments were run on 34 artifact-free subjects. We report three overlap and three surface distance metrics to evaluate the agreement between the automatic and manual segmentations, namely the dice similarity coefficient (DSC), recall (RE), precision (PR), average surface distance (ASD), root-mean-square surface distance (RMSSD), and Hausdorff distance (HD). Analysis of variances was used to compare pooled label metrics between the three algorithms and the DSC on a 'per-organ' basis. A Mann-Whitney U test was used to compare the pooled metrics between CFs and CNNs and the DSC on a 'per-organ' basis, when using different imaging combinations as input for training. All three algorithms resulted in robust segmenters that were effectively trained using a relatively small number of datasets, an important consideration in the clinical setting. Mean overlap metrics for all the segmented structures were: CFs: DSC = 0.70 ± 0.18, RE = 0.73 ± 0.18, PR = 0.71 ± 0.14, CNNs: DSC = 0.81 ± 0.13, RE = 0.83 ± 0.14, PR = 0.82 ± 0.10, MA: DSC = 0.71 ± 0.22, RE = 0.70 ± 0.34, PR = 0.77 ± 0.15. Mean surface distance

SC-CNNs for chaotic signal applications in secure communication systems.

Science.gov (United States)

Caponetto, Riccardo; Fortuna, Luigi; Occhipinti, Luigi; Xibilia, Maria Gabriella

2003-12-01

In this paper a CNNs based circuit for the generation of hyperchaotic signals is proposed. The circuit has been developed for applications in secure communication systems. An Saito oscillator has been designed by using a suitable configuration of a four-cells State-Controlled CNNs. A cryptography system based on the Saito oscillator has been implemented by using inverse system synchronization. The proposed circuit implementation and experimental results are given.

Deep CNNs Along the Time Axis With Intermap Pooling for Robustness to Spectral Variations

Science.gov (United States)

Lee, Hwaran; Kim, Geonmin; Kim, Ho-Gyeong; Oh, Sang-Hoon; Lee, Soo-Young

2016-10-01

Convolutional neural networks (CNNs) with convolutional and pooling operations along the frequency axis have been proposed to attain invariance to frequency shifts of features. However, this is inappropriate with regard to the fact that acoustic features vary in frequency. In this paper, we contend that convolution along the time axis is more effective. We also propose the addition of an intermap pooling (IMP) layer to deep CNNs. In this layer, filters in each group extract common but spectrally variant features, then the layer pools the feature maps of each group. As a result, the proposed IMP CNN can achieve insensitivity to spectral variations characteristic of different speakers and utterances. The effectiveness of the IMP CNN architecture is demonstrated on several LVCSR tasks. Even without speaker adaptation techniques, the architecture achieved a WER of 12.7% on the SWB part of the Hub5'2000 evaluation test set, which is competitive with other state-of-the-art methods.

Adaptive Graph Convolutional Neural Networks

OpenAIRE

Li, Ruoyu; Wang, Sheng; Zhu, Feiyun; Huang, Junzhou

2018-01-01

Graph Convolutional Neural Networks (Graph CNNs) are generalizations of classical CNNs to handle graph data such as molecular data, point could and social networks. Current filters in graph CNNs are built for fixed and shared graph structure. However, for most real data, the graph structures varies in both size and connectivity. The paper proposes a generalized and flexible graph CNN taking data of arbitrary graph structure as input. In that way a task-driven adaptive graph is learned for eac...

A hybrid model based on neural networks for biomedical relation extraction.

Science.gov (United States)

Zhang, Yijia; Lin, Hongfei; Yang, Zhihao; Wang, Jian; Zhang, Shaowu; Sun, Yuanyuan; Yang, Liang

2018-05-01

Biomedical relation extraction can automatically extract high-quality biomedical relations from biomedical texts, which is a vital step for the mining of biomedical knowledge hidden in the literature. Recurrent neural networks (RNNs) and convolutional neural networks (CNNs) are two major neural network models for biomedical relation extraction. Neural network-based methods for biomedical relation extraction typically focus on the sentence sequence and employ RNNs or CNNs to learn the latent features from sentence sequences separately. However, RNNs and CNNs have their own advantages for biomedical relation extraction. Combining RNNs and CNNs may improve biomedical relation extraction. In this paper, we present a hybrid model for the extraction of biomedical relations that combines RNNs and CNNs. First, the shortest dependency path (SDP) is generated based on the dependency graph of the candidate sentence. To make full use of the SDP, we divide the SDP into a dependency word sequence and a relation sequence. Then, RNNs and CNNs are employed to automatically learn the features from the sentence sequence and the dependency sequences, respectively. Finally, the output features of the RNNs and CNNs are combined to detect and extract biomedical relations. We evaluate our hybrid model using five public (protein-protein interaction) PPI corpora and a (drug-drug interaction) DDI corpus. The experimental results suggest that the advantages of RNNs and CNNs in biomedical relation extraction are complementary. Combining RNNs and CNNs can effectively boost biomedical relation extraction performance. Copyright © 2018 Elsevier Inc. All rights reserved.

A Comprehensive Study on Cross-View Gait Based Human Identification with Deep CNNs.

Science.gov (United States)

Wu, Zifeng; Huang, Yongzhen; Wang, Liang; Wang, Xiaogang; Tan, Tieniu

2017-02-01

This paper studies an approach to gait based human identification via similarity learning by deep convolutional neural networks (CNNs). With a pretty small group of labeled multi-view human walking videos, we can train deep networks to recognize the most discriminative changes of gait patterns which suggest the change of human identity. To the best of our knowledge, this is the first work based on deep CNNs for gait recognition in the literature. Here, we provide an extensive empirical evaluation in terms of various scenarios, namely, cross-view and cross-walking-condition, with different preprocessing approaches and network architectures. The method is first evaluated on the challenging CASIA-B dataset in terms of cross-view gait recognition. Experimental results show that it outperforms the previous state-of-the-art methods by a significant margin. In particular, our method shows advantages when the cross-view angle is large, i.e., no less than 36 degree. And the average recognition rate can reach 94 percent, much better than the previous best result (less than 65 percent). The method is further evaluated on the OU-ISIR gait dataset to test its generalization ability to larger data. OU-ISIR is currently the largest dataset available in the literature for gait recognition, with 4,007 subjects. On this dataset, the average accuracy of our method under identical view conditions is above 98 percent, and the one for cross-view scenarios is above 91 percent. Finally, the method also performs the best on the USF gait dataset, whose gait sequences are imaged in a real outdoor scene. These results show great potential of this method for practical applications.

Global exponential stability and existence of periodic solutions of CNNs with delays

Science.gov (United States)

Dong, Meifang

2002-07-01

In this Letter, we establish general sufficient conditions for global exponential stability and existence of periodic solutions of a class of cellular neural networks (CNNs) with delays. The key to proving the sufficient conditions is the construction of a new Lyapunov functional. An elementary inequality, which may be of independent interest, has been employed in the proof. Checking the sufficient conditions is often reduced to checking some algebraic relations among certain set of parameter. Our sufficient conditions recover the known results in literature as special cases. Finally, we give two examples to illustrate the usage of our main results.

Cellular Neural Network for Real Time Image Processing

International Nuclear Information System (INIS)

Vagliasindi, G.; Arena, P.; Fortuna, L.; Mazzitelli, G.; Murari, A.

2008-01-01

Since their introduction in 1988, Cellular Nonlinear Networks (CNNs) have found a key role as image processing instruments. Thanks to their structure they are able of processing individual pixels in a parallel way providing fast image processing capabilities that has been applied to a wide range of field among which nuclear fusion. In the last years, indeed, visible and infrared video cameras have become more and more important in tokamak fusion experiments for the twofold aim of understanding the physics and monitoring the safety of the operation. Examining the output of these cameras in real-time can provide significant information for plasma control and safety of the machines. The potentiality of CNNs can be exploited to this aim. To demonstrate the feasibility of the approach, CNN image processing has been applied to several tasks both at the Frascati Tokamak Upgrade (FTU) and the Joint European Torus (JET)

Building change detection via a combination of CNNs using only RGB aerial imageries

Science.gov (United States)

Nemoto, Keisuke; Hamaguchi, Ryuhei; Sato, Masakazu; Fujita, Aito; Imaizumi, Tomoyuki; Hikosaka, Shuhei

2017-10-01

Building change information extracted from remote sensing imageries is important for various applications such as urban management and marketing planning. The goal of this work is to develop a methodology for automatically capturing building changes from remote sensing imageries. Recent studies have addressed this goal by exploiting 3-D information as a proxy for building height. In contrast, because in practice it is expensive or impossible to prepare 3-D information, we do not rely on 3-D data but focus on using only RGB aerial imageries. Instead, we employ deep convolutional neural networks (CNNs) to extract effective features, and improve change detection accuracy in RGB remote sensing imageries. We consider two aspects of building change detection, building detection and subsequent change detection. Our proposed methodology was tested on several areas, which has some differences such as dominant building characteristics and varying brightness values. On all over the tested areas, the proposed method provides good results for changed objects, with recall values over 75 % with a strict overlap requirement of over 50% in intersection-over-union (IoU). When the IoU threshold was relaxed to over 10%, resulting recall values were over 81%. We conclude that use of CNNs enables accurate detection of building changes without employing 3-D information.

Interleaved 3D-CNNs for joint segmentation of small-volume structures in head and neck CT images.

Science.gov (United States)

Ren, Xuhua; Xiang, Lei; Nie, Dong; Shao, Yeqin; Zhang, Huan; Shen, Dinggang; Wang, Qian

2018-02-26

Accurate 3D image segmentation is a crucial step in radiation therapy planning of head and neck tumors. These segmentation results are currently obtained by manual outlining of tissues, which is a tedious and time-consuming procedure. Automatic segmentation provides an alternative solution, which, however, is often difficult for small tissues (i.e., chiasm and optic nerves in head and neck CT images) because of their small volumes and highly diverse appearance/shape information. In this work, we propose to interleave multiple 3D Convolutional Neural Networks (3D-CNNs) to attain automatic segmentation of small tissues in head and neck CT images. A 3D-CNN was designed to segment each structure of interest. To make full use of the image appearance information, multiscale patches are extracted to describe the center voxel under consideration and then input to the CNN architecture. Next, as neighboring tissues are often highly related in the physiological and anatomical perspectives, we interleave the CNNs designated for the individual tissues. In this way, the tentative segmentation result of a specific tissue can contribute to refine the segmentations of other neighboring tissues. Finally, as more CNNs are interleaved and cascaded, a complex network of CNNs can be derived, such that all tissues can be jointly segmented and iteratively refined. Our method was validated on a set of 48 CT images, obtained from the Medical Image Computing and Computer Assisted Intervention (MICCAI) Challenge 2015. The Dice coefficient (DC) and the 95% Hausdorff Distance (95HD) are computed to measure the accuracy of the segmentation results. The proposed method achieves higher segmentation accuracy (with the average DC: 0.58 ± 0.17 for optic chiasm, and 0.71 ± 0.08 for optic nerve; 95HD: 2.81 ± 1.56 mm for optic chiasm, and 2.23 ± 0.90 mm for optic nerve) than the MICCAI challenge winner (with the average DC: 0.38 for optic chiasm, and 0.68 for optic nerve; 95HD: 3.48 for

Detector development and background estimation for the observation of Coherent Neutrino Nucleus Scattering (CNNS)

Energy Technology Data Exchange (ETDEWEB)

Guetlein, Achim; Ciemniak, Christian; Feilitzsch, Franz von; Lanfranchi, Jean-Come; Oberauer, Lothar; Potzel, Walter; Roth, Sabine; Schoenert, Stefan; Sivers, Moritz von; Strauss, Raimund; Wawoczny, Stefan; Willers, Michael; Zoeller, Andreas [Technische Universitaet Muenchen, Physik-Department, E15 (Germany)

2012-07-01

The Coherent Neutrino Nucleus Scattering (CNNS) is a neutral current process of the weak interaction and is thus flavor independent. A low-energetic neutrino scatters off a target nucleus. For low transferred momenta the wavelength of the transferred Z{sup 0} boson is comparable to the diameter of the target nucleus. Thus, the neutrino interacts with all nucleons coherently and the cross section for the CNNS is enhanced. To observe CNNS for the first time we are developing cryogenic detectors with a target mass of about 10 g each and an energy threshold of less than 0.5 keV. The current status of this development is presented as well as the estimated background for an experiment in the vicinity of a nuclear power reactor as a strong neutrino source.

Can higher education improve the professional identity of CNNs?

Science.gov (United States)

Frankland, Sharon

2010-12-01

Community nursery nurses (CNNs) are an important part of the multidisciplinary team. This paper discusses how two students undertaking a foundation degree in early years experienced changes to their personal and professional identities. A 'life history' approach was used to understand and interpret their experiences in depth. Both students would not have entered higher education had it not been for the widening participation drive. The higher education experience had a positive influence on personal and professional identity for the learners. They underlined that the widening participation drive can enable those students from 'non-traditional backgrounds' to enter and benefit from higher education. However, changes to higher education funding and public sector cutbacks have raised grave concerns about the continued ability of CNNs and other early years practitioners to access such courses. This could have a negative effect on the continuing professional development and subsequent changing identities within this particular group.

Neural Networks for Non-linear Control

DEFF Research Database (Denmark)

Sørensen, O.

1994-01-01

This paper describes how a neural network, structured as a Multi Layer Perceptron, is trained to predict, simulate and control a non-linear process.......This paper describes how a neural network, structured as a Multi Layer Perceptron, is trained to predict, simulate and control a non-linear process....

Ethical conflicts reported by Certified Nephrology Nurses (CNNs) practicing in dialysis settings.

Science.gov (United States)

Redman, B K; Hill, M N; Fry, S T

1997-02-01

The purpose of the investigators was to describe and classify ethical conflicts experienced by Certified Nephrology Nurses (CNNs) practicing in dialysis settings in four eastern states and the District of Columbia, and to explore associated demographic, educational, and practice setting factors associated with these ethical conflicts. A descriptive survey design was used. All members of the American Nephrology Nurses Association (ANNA) who were CNNs working in Maryland, Virginia, and the District of Columbia, and a random sample of those in New York State and Pennsylvania were contacted. Those CNNs working in dialysis settings were asked to complete the Demographic Data Form and the Moral Conflict Questionnaire developed by Fry (1990). Eighty-eight met inclusion criteria, agreed to participate, and described ethical conflicts. By far the most common practice context for the described ethical conflicts were decisions about discontinuation or initiation of dialysis (69%). Participants were clear about the moral problem and ethical principles involved. Participants reported being involved in serious ethical conflicts about patient care. Since two-thirds were not resolved, further research should investigate whether existing mechanisms in practice settings for resolution of ethical conflicts are not working or are not being used by nurses.

Tuning chaos in network sharing common nonlinearity

Science.gov (United States)

Paul Asir, M.; Jeevarekha, A.; Philominathan, P.

2016-06-01

In this paper, a novel type of network called network sharing common nonlinearity comprising both autonomous and non-autonomous oscillators have been investigated. We propose that these networks are robust for operating at desired modes i.e., chaotic or periodic by altering the v-i characteristics of common nonlinear element alone. The dynamics of these networks were examined through numerical, analytical, experimental and Multisim simulations.

Nonlinear threshold Boolean automata networks and phase transitions

OpenAIRE

Demongeot, Jacques; Sené, Sylvain

2010-01-01

In this report, we present a formal approach that addresses the problem of emergence of phase transitions in stochastic and attractive nonlinear threshold Boolean automata networks. Nonlinear networks considered are informally defined on the basis of classical stochastic threshold Boolean automata networks in which specific interaction potentials of neighbourhood coalition are taken into account. More precisely, specific nonlinear terms compose local transition functions that define locally t...

Classification of MR brain images by combination of multi-CNNs for AD diagnosis

Science.gov (United States)

Cheng, Danni; Liu, Manhua; Fu, Jianliang; Wang, Yaping

2017-07-01

Alzheimer's disease (AD) is an irreversible neurodegenerative disorder with progressive impairment of memory and cognitive functions. Its early diagnosis is crucial for development of future treatment. Magnetic resonance images (MRI) play important role to help understand the brain anatomical changes related to AD. Conventional methods extract the hand-crafted features such as gray matter volumes and cortical thickness and train a classifier to distinguish AD from other groups. Different from these methods, this paper proposes to construct multiple deep 3D convolutional neural networks (3D-CNNs) to learn the various features from local brain images which are combined to make the final classification for AD diagnosis. First, a number of local image patches are extracted from the whole brain image and a 3D-CNN is built upon each local patch to transform the local image into more compact high-level features. Then, the upper convolution and fully connected layers are fine-tuned to combine the multiple 3D-CNNs for image classification. The proposed method can automatically learn the generic features from imaging data for classification. Our method is evaluated using T1-weighted structural MR brain images on 428 subjects including 199 AD patients and 229 normal controls (NC) from Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Experimental results show that the proposed method achieves an accuracy of 87.15% and an AUC (area under the ROC curve) of 92.26% for AD classification, demonstrating the promising classification performances.

User-generated content curation with deep convolutional neural networks

OpenAIRE

Tous Liesa, Rubén; Wust, Otto; Gómez, Mauro; Poveda, Jonatan; Elena, Marc; Torres Viñals, Jordi; Makni, Mouna; Ayguadé Parra, Eduard

2016-01-01

In this paper, we report a work consisting in using deep convolutional neural networks (CNNs) for curating and filtering photos posted by social media users (Instagram and Twitter). The final goal is to facilitate searching and discovering user-generated content (UGC) with potential value for digital marketing tasks. The images are captured in real time and automatically annotated with multiple CNNs. Some of the CNNs perform generic object recognition tasks while others perform what we call v...

Single Image Super-Resolution Based on Multi-Scale Competitive Convolutional Neural Network.

Science.gov (United States)

Du, Xiaofeng; Qu, Xiaobo; He, Yifan; Guo, Di

2018-03-06

Deep convolutional neural networks (CNNs) are successful in single-image super-resolution. Traditional CNNs are limited to exploit multi-scale contextual information for image reconstruction due to the fixed convolutional kernel in their building modules. To restore various scales of image details, we enhance the multi-scale inference capability of CNNs by introducing competition among multi-scale convolutional filters, and build up a shallow network under limited computational resources. The proposed network has the following two advantages: (1) the multi-scale convolutional kernel provides the multi-context for image super-resolution, and (2) the maximum competitive strategy adaptively chooses the optimal scale of information for image reconstruction. Our experimental results on image super-resolution show that the performance of the proposed network outperforms the state-of-the-art methods.

Multilevel Contextual 3-D CNNs for False Positive Reduction in Pulmonary Nodule Detection.

Science.gov (United States)

Dou, Qi; Chen, Hao; Yu, Lequan; Qin, Jing; Heng, Pheng-Ann

2017-07-01

False positive reduction is one of the most crucial components in an automated pulmonary nodule detection system, which plays an important role in lung cancer diagnosis and early treatment. The objective of this paper is to effectively address the challenges in this task and therefore to accurately discriminate the true nodules from a large number of candidates. We propose a novel method employing three-dimensional (3-D) convolutional neural networks (CNNs) for false positive reduction in automated pulmonary nodule detection from volumetric computed tomography (CT) scans. Compared with its 2-D counterparts, the 3-D CNNs can encode richer spatial information and extract more representative features via their hierarchical architecture trained with 3-D samples. More importantly, we further propose a simple yet effective strategy to encode multilevel contextual information to meet the challenges coming with the large variations and hard mimics of pulmonary nodules. The proposed framework has been extensively validated in the LUNA16 challenge held in conjunction with ISBI 2016, where we achieved the highest competition performance metric (CPM) score in the false positive reduction track. Experimental results demonstrated the importance and effectiveness of integrating multilevel contextual information into 3-D CNN framework for automated pulmonary nodule detection in volumetric CT data. While our method is tailored for pulmonary nodule detection, the proposed framework is general and can be easily extended to many other 3-D object detection tasks from volumetric medical images, where the targeting objects have large variations and are accompanied by a number of hard mimics.

Cell dynamic morphology classification using deep convolutional neural networks.

Science.gov (United States)

Li, Heng; Pang, Fengqian; Shi, Yonggang; Liu, Zhiwen

2018-05-15

Cell morphology is often used as a proxy measurement of cell status to understand cell physiology. Hence, interpretation of cell dynamic morphology is a meaningful task in biomedical research. Inspired by the recent success of deep learning, we here explore the application of convolutional neural networks (CNNs) to cell dynamic morphology classification. An innovative strategy for the implementation of CNNs is introduced in this study. Mouse lymphocytes were collected to observe the dynamic morphology, and two datasets were thus set up to investigate the performances of CNNs. Considering the installation of deep learning, the classification problem was simplified from video data to image data, and was then solved by CNNs in a self-taught manner with the generated image data. CNNs were separately performed in three installation scenarios and compared with existing methods. Experimental results demonstrated the potential of CNNs in cell dynamic morphology classification, and validated the effectiveness of the proposed strategy. CNNs were successfully applied to the classification problem, and outperformed the existing methods in the classification accuracy. For the installation of CNNs, transfer learning was proved to be a promising scheme. © 2018 International Society for Advancement of Cytometry. © 2018 International Society for Advancement of Cytometry.

Forecasting Flare Activity Using Deep Convolutional Neural Networks

Science.gov (United States)

Hernandez, T.

2017-12-01

Current operational flare forecasting relies on human morphological analysis of active regions and the persistence of solar flare activity through time (i.e. that the Sun will continue to do what it is doing right now: flaring or remaining calm). In this talk we present the results of applying deep Convolutional Neural Networks (CNNs) to the problem of solar flare forecasting. CNNs operate by training a set of tunable spatial filters that, in combination with neural layer interconnectivity, allow CNNs to automatically identify significant spatial structures predictive for classification and regression problems. We will start by discussing the applicability and success rate of the approach, the advantages it has over non-automated forecasts, and how mining our trained neural network provides a fresh look into the mechanisms behind magnetic energy storage and release.

Finding Neutrinos in LArTPCs using Convolutional Neural Networks

Science.gov (United States)

Wongjirad, Taritree

2017-09-01

Deep learning algorithms, which have emerged over the last decade, are opening up new ways to analyze data for many particle physics experiments. MicroBooNE, which is a neutrino experiment at Fermilab, has been exploring the use of such algorithms, in particular, convolutional neural networks (CNNS). CNNs are the state-of-the-art method for a large class of problems involving the analysis of images. This makes CNNs an attractive approach for MicroBooNE, whose detector, a liquid argon time projection chamber (LArTPC), produces high-resolution images of particle interactions. In this talk, I will discuss the ways CNNs can be applied to tasks like neutrino interaction detection and particle identification in MicroBooNE and LArTPCs.

Linking structure and activity in nonlinear spiking networks.

Directory of Open Access Journals (Sweden)

Gabriel Koch Ocker

2017-06-01

Full Text Available Recent experimental advances are producing an avalanche of data on both neural connectivity and neural activity. To take full advantage of these two emerging datasets we need a framework that links them, revealing how collective neural activity arises from the structure of neural connectivity and intrinsic neural dynamics. This problem of structure-driven activity has drawn major interest in computational neuroscience. Existing methods for relating activity and architecture in spiking networks rely on linearizing activity around a central operating point and thus fail to capture the nonlinear responses of individual neurons that are the hallmark of neural information processing. Here, we overcome this limitation and present a new relationship between connectivity and activity in networks of nonlinear spiking neurons by developing a diagrammatic fluctuation expansion based on statistical field theory. We explicitly show how recurrent network structure produces pairwise and higher-order correlated activity, and how nonlinearities impact the networks' spiking activity. Our findings open new avenues to investigating how single-neuron nonlinearities-including those of different cell types-combine with connectivity to shape population activity and function.

Linking structure and activity in nonlinear spiking networks.

Science.gov (United States)

Ocker, Gabriel Koch; Josić, Krešimir; Shea-Brown, Eric; Buice, Michael A

2017-06-01

Recent experimental advances are producing an avalanche of data on both neural connectivity and neural activity. To take full advantage of these two emerging datasets we need a framework that links them, revealing how collective neural activity arises from the structure of neural connectivity and intrinsic neural dynamics. This problem of structure-driven activity has drawn major interest in computational neuroscience. Existing methods for relating activity and architecture in spiking networks rely on linearizing activity around a central operating point and thus fail to capture the nonlinear responses of individual neurons that are the hallmark of neural information processing. Here, we overcome this limitation and present a new relationship between connectivity and activity in networks of nonlinear spiking neurons by developing a diagrammatic fluctuation expansion based on statistical field theory. We explicitly show how recurrent network structure produces pairwise and higher-order correlated activity, and how nonlinearities impact the networks' spiking activity. Our findings open new avenues to investigating how single-neuron nonlinearities-including those of different cell types-combine with connectivity to shape population activity and function.

Image processing with a cellular nonlinear network

International Nuclear Information System (INIS)

Morfu, S.

2005-01-01

A cellular nonlinear network (CNN) based on uncoupled nonlinear oscillators is proposed for image processing purposes. It is shown theoretically and numerically that the contrast of an image loaded at the nodes of the CNN is strongly enhanced, even if this one is initially weak. An image inversion can be also obtained without reconfiguration of the network whereas a gray levels extraction can be performed with an additional threshold filtering. Lastly, an electronic implementation of this CNN is presented

Cellular Neural Networks: A genetic algorithm for parameters optimization in artificial vision applications

Energy Technology Data Exchange (ETDEWEB)

Taraglio, S. [ENEA, Centro Ricerche Casaccia, Rome (Italy). Dipt. Innovazione; Zanela, A. [Rome Univ. `La Sapienza` (Italy). Dipt. di Fisica

1997-03-01

An optimization method for some of the CNN`s (Cellular Neural Network) parameters, based on evolutionary strategies, is proposed. The new class of feedback template found is more effective in extracting features from the images that an autonomous vehicle acquires, than in the previous CNN`s literature.

Network science, nonlinear science and infrastructure systems

CERN Document Server

2007-01-01

Network Science, Nonlinear Science and Infrastructure Systems has been written by leading scholars in these areas. Its express purpose is to develop common theoretical underpinnings to better solve modern infrastructural problems. It is felt by many who work in these fields that many modern communication problems, ranging from transportation networks to telecommunications, Internet, supply chains, etc., are fundamentally infrastructure problems. Moreover, these infrastructure problems would benefit greatly from a confluence of theoretical and methodological work done with the areas of Network Science, Dynamical Systems and Nonlinear Science. This book is dedicated to the formulation of infrastructural tools that will better solve these types of infrastructural problems. .

Identification of Non-Linear Structures using Recurrent Neural Networks

DEFF Research Database (Denmark)

Kirkegaard, Poul Henning; Nielsen, Søren R. K.; Hansen, H. I.

Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure.......Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure....

Identification of Non-Linear Structures using Recurrent Neural Networks

DEFF Research Database (Denmark)

Kirkegaard, Poul Henning; Nielsen, Søren R. K.; Hansen, H. I.

1995-01-01

Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure.......Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure....

Adaptive nonlinear control using input normalized neural networks

International Nuclear Information System (INIS)

Leeghim, Henzeh; Seo, In Ho; Bang, Hyo Choong

2008-01-01

An adaptive feedback linearization technique combined with the neural network is addressed to control uncertain nonlinear systems. The neural network-based adaptive control theory has been widely studied. However, the stability analysis of the closed-loop system with the neural network is rather complicated and difficult to understand, and sometimes unnecessary assumptions are involved. As a result, unnecessary assumptions for stability analysis are avoided by using the neural network with input normalization technique. The ultimate boundedness of the tracking error is simply proved by the Lyapunov stability theory. A new simple update law as an adaptive nonlinear control is derived by the simplification of the input normalized neural network assuming the variation of the uncertain term is sufficiently small

Automatic Seismic-Event Classification with Convolutional Neural Networks.

Science.gov (United States)

Bueno Rodriguez, A.; Titos Luzón, M.; Garcia Martinez, L.; Benitez, C.; Ibáñez, J. M.

2017-12-01

Active volcanoes exhibit a wide range of seismic signals, providing vast amounts of unlabelled volcano-seismic data that can be analyzed through the lens of artificial intelligence. However, obtaining high-quality labelled data is time-consuming and expensive. Deep neural networks can process data in their raw form, compute high-level features and provide a better representation of the input data distribution. These systems can be deployed to classify seismic data at scale, enhance current early-warning systems and build extensive seismic catalogs. In this research, we aim to classify spectrograms from seven different seismic events registered at "Volcán de Fuego" (Colima, Mexico), during four eruptive periods. Our approach is based on convolutional neural networks (CNNs), a sub-type of deep neural networks that can exploit grid structure from the data. Volcano-seismic signals can be mapped into a grid-like structure using the spectrogram: a representation of the temporal evolution in terms of time and frequency. Spectrograms were computed from the data using Hamming windows with 4 seconds length, 2.5 seconds overlapping and 128 points FFT resolution. Results are compared to deep neural networks, random forest and SVMs. Experiments show that CNNs can exploit temporal and frequency information, attaining a classification accuracy of 93%, similar to deep networks 91% but outperforming SVM and random forest. These results empirically show that CNNs are powerful models to classify a wide range of volcano-seismic signals, and achieve good generalization. Furthermore, volcano-seismic spectrograms contains useful discriminative information for the CNN, as higher layers of the network combine high-level features computed for each frequency band, helping to detect simultaneous events in time. Being at the intersection of deep learning and geophysics, this research enables future studies of how CNNs can be used in volcano monitoring to accurately determine the detection and

Nonlinearity and chaos in wireless network traffic

International Nuclear Information System (INIS)

Mukherjee, Somenath; Ray, Rajdeep; Samanta, Rajkumar; Khondekar, Mofazzal H.; Sanyal, Goutam

2017-01-01

The natural complexity of wireless mobile network traffic dynamics has been assessed in this article by tracing the presence of nonlinearity and chaos in the profile of daily peak hour call arrival and daily call drop of a sub-urban local mobile switching centre. The tools like Recurrence Plot and Recurrence Quantification Analysis (RQA) has been used to reveal the probable presence of non-stationarity, nonlinearity and chaosity in the network traffic. Information Entropy (IE) and 0–1 test have been employed to provide the quantitative support to the findings. Both the daily peak hour call arrival profile and the daily call drop profile exhibit non-stationarity, determinism and nonlinearity with the former one being more regular while the later one is chaotic.

Robustness Design for CNN Templates with Performance of Extracting Closed Domain

International Nuclear Information System (INIS)

Li Weidong; Min Lequan

2006-01-01

The cellular neural/nonlinear network (CNN) is a powerful tool for image and video signal processing, robotic and biological visions. This paper introduces a kind of CNNs with performance of extracting closed domains in binary images, and gives a general method for designing templates of such a kind of CNNs. One theorem provides parameter inequalities for determining parameter intervals for implementing prescribed image processing functions, respectively. Examples for extracting closed domains in binary scale images are given.

Applying Gradient Descent in Convolutional Neural Networks

Science.gov (United States)

Cui, Nan

2018-04-01

With the development of the integrated circuit and computer science, people become caring more about solving practical issues via information technologies. Along with that, a new subject called Artificial Intelligent (AI) comes up. One popular research interest of AI is about recognition algorithm. In this paper, one of the most common algorithms, Convolutional Neural Networks (CNNs) will be introduced, for image recognition. Understanding its theory and structure is of great significance for every scholar who is interested in this field. Convolution Neural Network is an artificial neural network which combines the mathematical method of convolution and neural network. The hieratical structure of CNN provides it reliable computer speed and reasonable error rate. The most significant characteristics of CNNs are feature extraction, weight sharing and dimension reduction. Meanwhile, combining with the Back Propagation (BP) mechanism and the Gradient Descent (GD) method, CNNs has the ability to self-study and in-depth learning. Basically, BP provides an opportunity for backwardfeedback for enhancing reliability and GD is used for self-training process. This paper mainly discusses the CNN and the related BP and GD algorithms, including the basic structure and function of CNN, details of each layer, the principles and features of BP and GD, and some examples in practice with a summary in the end.

Network evolution by nonlinear preferential rewiring of edges

Science.gov (United States)

Xu, Xin-Jian; Hu, Xiao-Ming; Zhang, Li-Jie

2011-06-01

The mathematical framework for small-world networks proposed in a seminal paper by Watts and Strogatz sparked a widespread interest in modeling complex networks in the past decade. However, most of research contributing to static models is in contrast to real-world dynamic networks, such as social and biological networks, which are characterized by rearrangements of connections among agents. In this paper, we study dynamic networks evolved by nonlinear preferential rewiring of edges. The total numbers of vertices and edges of the network are conserved, but edges are continuously rewired according to the nonlinear preference. Assuming power-law kernels with exponents α and β, the network structures in stationary states display a distinct behavior, depending only on β. For β>1, the network is highly heterogeneous with the emergence of starlike structures. For β<1, the network is widely homogeneous with a typical connectivity. At β=1, the network is scale free with an exponential cutoff.

Transfer Learning with Convolutional Neural Networks for SAR Ship Recognition

Science.gov (United States)

Zhang, Di; Liu, Jia; Heng, Wang; Ren, Kaijun; Song, Junqiang

2018-03-01

Ship recognition is the backbone of marine surveillance systems. Recent deep learning methods, e.g. Convolutional Neural Networks (CNNs), have shown high performance for optical images. Learning CNNs, however, requires a number of annotated samples to estimate numerous model parameters, which prevents its application to Synthetic Aperture Radar (SAR) images due to the limited annotated training samples. Transfer learning has been a promising technique for applications with limited data. To this end, a novel SAR ship recognition method based on CNNs with transfer learning has been developed. In this work, we firstly start with a CNNs model that has been trained in advance on Moving and Stationary Target Acquisition and Recognition (MSTAR) database. Next, based on the knowledge gained from this image recognition task, we fine-tune the CNNs on a new task to recognize three types of ships in the OpenSARShip database. The experimental results show that our proposed approach can obviously increase the recognition rate comparing with the result of merely applying CNNs. In addition, compared to existing methods, the proposed method proves to be very competitive and can learn discriminative features directly from training data instead of requiring pre-specification or pre-selection manually.

Robust receding horizon control for networked and distributed nonlinear systems

CERN Document Server

Li, Huiping

2017-01-01

This book offers a comprehensive, easy-to-understand overview of receding-horizon control for nonlinear networks. It presents novel general strategies that can simultaneously handle general nonlinear dynamics, system constraints, and disturbances arising in networked and large-scale systems and which can be widely applied. These receding-horizon-control-based strategies can achieve sub-optimal control performance while ensuring closed-loop stability: a feature attractive to engineers. The authors address the problems of networked and distributed control step-by-step, gradually increasing the level of challenge presented. The book first introduces the state-feedback control problems of nonlinear networked systems and then studies output feedback control problems. For large-scale nonlinear systems, disturbance is considered first, then communication delay separately, and lastly the simultaneous combination of delays and disturbances. Each chapter of this easy-to-follow book not only proposes and analyzes novel ...

Delay-dependent exponential stability of cellular neural networks with time-varying delays

International Nuclear Information System (INIS)

Zhang Qiang; Wei Xiaopeng; Xu Jin

2005-01-01

The global exponential stability of cellular neural networks (CNNs) with time-varying delays is analyzed. Two new sufficient conditions ensuring global exponential stability for delayed CNNs are obtained. The conditions presented here are related to the size of delay. The stability results improve the earlier publications. Two examples are given to demonstrate the effectiveness of the obtained results

Consensus problem in directed networks of multi-agents via nonlinear protocols

International Nuclear Information System (INIS)

Liu Xiwei; Chen Tianping; Lu Wenlian

2009-01-01

In this Letter, the consensus problem via distributed nonlinear protocols for directed networks is investigated. Its dynamical behaviors are described by ordinary differential equations (ODEs). Based on graph theory, matrix theory and the Lyapunov direct method, some sufficient conditions of nonlinear protocols guaranteeing asymptotical or exponential consensus are presented and rigorously proved. The main contribution of this work is that for nonlinearly coupled networks, we generalize the results for undirected networks to directed networks. Consensus under pinning control technique is also developed here. Simulations are also given to show the validity of the theories.

A Recurrent Neural Network for Nonlinear Fractional Programming

Directory of Open Access Journals (Sweden)

Quan-Ju Zhang

2012-01-01

Full Text Available This paper presents a novel recurrent time continuous neural network model which performs nonlinear fractional optimization subject to interval constraints on each of the optimization variables. The network is proved to be complete in the sense that the set of optima of the objective function to be minimized with interval constraints coincides with the set of equilibria of the neural network. It is also shown that the network is primal and globally convergent in the sense that its trajectory cannot escape from the feasible region and will converge to an exact optimal solution for any initial point being chosen in the feasible interval region. Simulation results are given to demonstrate further the global convergence and good performance of the proposing neural network for nonlinear fractional programming problems with interval constraints.

Nonlinear programming with feedforward neural networks.

Energy Technology Data Exchange (ETDEWEB)

Reifman, J.

1999-06-02

We provide a practical and effective method for solving constrained optimization problems by successively training a multilayer feedforward neural network in a coupled neural-network/objective-function representation. Nonlinear programming problems are easily mapped into this representation which has a simpler and more transparent method of solution than optimization performed with Hopfield-like networks and poses very mild requirements on the functions appearing in the problem. Simulation results are illustrated and compared with an off-the-shelf optimization tool.

Nonlinear Time Series Prediction Using Chaotic Neural Networks

Science.gov (United States)

Li, Ke-Ping; Chen, Tian-Lun

2001-06-01

A nonlinear feedback term is introduced into the evaluation equation of weights of the backpropagation algorithm for neural network, the network becomes a chaotic one. For the purpose of that we can investigate how the different feedback terms affect the process of learning and forecasting, we use the model to forecast the nonlinear time series which is produced by Makey-Glass equation. By selecting the suitable feedback term, the system can escape from the local minima and converge to the global minimum or its approximate solutions, and the forecasting results are better than those of backpropagation algorithm. The project supported by National Basic Research Project "Nonlinear Science" and National Natural Science Foundation of China under Grant No. 60074020

Do Convolutional Neural Networks Learn Class Hierarchy?

Science.gov (United States)

Bilal, Alsallakh; Jourabloo, Amin; Ye, Mao; Liu, Xiaoming; Ren, Liu

2018-01-01

Convolutional Neural Networks (CNNs) currently achieve state-of-the-art accuracy in image classification. With a growing number of classes, the accuracy usually drops as the possibilities of confusion increase. Interestingly, the class confusion patterns follow a hierarchical structure over the classes. We present visual-analytics methods to reveal and analyze this hierarchy of similar classes in relation with CNN-internal data. We found that this hierarchy not only dictates the confusion patterns between the classes, it furthermore dictates the learning behavior of CNNs. In particular, the early layers in these networks develop feature detectors that can separate high-level groups of classes quite well, even after a few training epochs. In contrast, the latter layers require substantially more epochs to develop specialized feature detectors that can separate individual classes. We demonstrate how these insights are key to significant improvement in accuracy by designing hierarchy-aware CNNs that accelerate model convergence and alleviate overfitting. We further demonstrate how our methods help in identifying various quality issues in the training data.

Distributed Extreme Learning Machine for Nonlinear Learning over Network

Directory of Open Access Journals (Sweden)

Songyan Huang

2015-02-01

Full Text Available Distributed data collection and analysis over a network are ubiquitous, especially over a wireless sensor network (WSN. To our knowledge, the data model used in most of the distributed algorithms is linear. However, in real applications, the linearity of systems is not always guaranteed. In nonlinear cases, the single hidden layer feedforward neural network (SLFN with radial basis function (RBF hidden neurons has the ability to approximate any continuous functions and, thus, may be used as the nonlinear learning system. However, confined by the communication cost, using the distributed version of the conventional algorithms to train the neural network directly is usually prohibited. Fortunately, based on the theorems provided in the extreme learning machine (ELM literature, we only need to compute the output weights of the SLFN. Computing the output weights itself is a linear learning problem, although the input-output mapping of the overall SLFN is still nonlinear. Using the distributed algorithmto cooperatively compute the output weights of the SLFN, we obtain a distributed extreme learning machine (dELM for nonlinear learning in this paper. This dELM is applied to the regression problem and classification problem to demonstrate its effectiveness and advantages.

A convolutional neural network neutrino event classifier

International Nuclear Information System (INIS)

Aurisano, A.; Sousa, A.; Radovic, A.; Vahle, P.; Rocco, D.; Pawloski, G.; Himmel, A.; Niner, E.; Messier, M.D.; Psihas, F.

2016-01-01

Convolutional neural networks (CNNs) have been widely applied in the computer vision community to solve complex problems in image recognition and analysis. We describe an application of the CNN technology to the problem of identifying particle interactions in sampling calorimeters used commonly in high energy physics and high energy neutrino physics in particular. Following a discussion of the core concepts of CNNs and recent innovations in CNN architectures related to the field of deep learning, we outline a specific application to the NOvA neutrino detector. This algorithm, CVN (Convolutional Visual Network) identifies neutrino interactions based on their topology without the need for detailed reconstruction and outperforms algorithms currently in use by the NOvA collaboration.

Optimization of a fluorescence X-ray source and background studies for a prospective CNNS experiment

Energy Technology Data Exchange (ETDEWEB)

Zoeller, Andreas; Ciemniak, Christian; Feilitzsch, Franz von; Guetlein, Achim; Haag, Nils; Hofmann, Martin; Isaila, Christian; Lanfranchi, Jean-Come; Oberauer, Lothar; Pfister, Sebastian; Potzel, Walter; Roth, Sabine; Schoenert, Stefan; Sivers, Moritz von; Strauss, Raimund [Technische Universitaet Muenchen, Physik-Department, E15 (Germany); Lachenmaier, Tobias [Eberhard Karls Universitaet, Tuebingen (Germany)

2011-07-01

Coherent Neutrino Nucleus Scattering (CNNS) is predicted by the Standard Model but hasn't been measured yet. A good background discrimination and shielding is essential for the achievement of a prospective experiment. We show the results of simulations for background discrimination and suppression using a myon veto system in combination with a shielding around a cryostat. With CNNS, the expected nuclear recoil energy for reactor anti-neutrinos is in the range of

EMG-Based Estimation of Limb Movement Using Deep Learning With Recurrent Convolutional Neural Networks.

Science.gov (United States)

Xia, Peng; Hu, Jie; Peng, Yinghong

2017-10-25

A novel model based on deep learning is proposed to estimate kinematic information for myoelectric control from multi-channel electromyogram (EMG) signals. The neural information of limb movement is embedded in EMG signals that are influenced by all kinds of factors. In order to overcome the negative effects of variability in signals, the proposed model employs the deep architecture combining convolutional neural networks (CNNs) and recurrent neural networks (RNNs). The EMG signals are transformed to time-frequency frames as the input to the model. The limb movement is estimated by the model that is trained with the gradient descent and backpropagation procedure. We tested the model for simultaneous and proportional estimation of limb movement in eight healthy subjects and compared it with support vector regression (SVR) and CNNs on the same data set. The experimental studies show that the proposed model has higher estimation accuracy and better robustness with respect to time. The combination of CNNs and RNNs can improve the model performance compared with using CNNs alone. The model of deep architecture is promising in EMG decoding and optimization of network structures can increase the accuracy and robustness. © 2017 International Center for Artificial Organs and Transplantation and Wiley Periodicals, Inc.

Nonlinear adaptive inverse control via the unified model neural network

Science.gov (United States)

Jeng, Jin-Tsong; Lee, Tsu-Tian

1999-03-01

In this paper, we propose a new nonlinear adaptive inverse control via a unified model neural network. In order to overcome nonsystematic design and long training time in nonlinear adaptive inverse control, we propose the approximate transformable technique to obtain a Chebyshev Polynomials Based Unified Model (CPBUM) neural network for the feedforward/recurrent neural networks. It turns out that the proposed method can use less training time to get an inverse model. Finally, we apply this proposed method to control magnetic bearing system. The experimental results show that the proposed nonlinear adaptive inverse control architecture provides a greater flexibility and better performance in controlling magnetic bearing systems.

Simulation of nonlinear random vibrations using artificial neural networks

Energy Technology Data Exchange (ETDEWEB)

Paez, T.L.; Tucker, S.; O`Gorman, C.

1997-02-01

The simulation of mechanical system random vibrations is important in structural dynamics, but it is particularly difficult when the system under consideration is nonlinear. Artificial neural networks provide a useful tool for the modeling of nonlinear systems, however, such modeling may be inefficient or insufficiently accurate when the system under consideration is complex. This paper shows that there are several transformations that can be used to uncouple and simplify the components of motion of a complex nonlinear system, thereby making its modeling and random vibration simulation, via component modeling with artificial neural networks, a much simpler problem. A numerical example is presented.

Deep Convolutional Neural Networks: Structure, Feature Extraction and Training

Directory of Open Access Journals (Sweden)

Namatēvs Ivars

2017-12-01

Full Text Available Deep convolutional neural networks (CNNs are aimed at processing data that have a known network like topology. They are widely used to recognise objects in images and diagnose patterns in time series data as well as in sensor data classification. The aim of the paper is to present theoretical and practical aspects of deep CNNs in terms of convolution operation, typical layers and basic methods to be used for training and learning. Some practical applications are included for signal and image classification. Finally, the present paper describes the proposed block structure of CNN for classifying crucial features from 3D sensor data.

Nonlinear signal processing using neural networks: Prediction and system modelling

Energy Technology Data Exchange (ETDEWEB)

Lapedes, A.; Farber, R.

1987-06-01

The backpropagation learning algorithm for neural networks is developed into a formalism for nonlinear signal processing. We illustrate the method by selecting two common topics in signal processing, prediction and system modelling, and show that nonlinear applications can be handled extremely well by using neural networks. The formalism is a natural, nonlinear extension of the linear Least Mean Squares algorithm commonly used in adaptive signal processing. Simulations are presented that document the additional performance achieved by using nonlinear neural networks. First, we demonstrate that the formalism may be used to predict points in a highly chaotic time series with orders of magnitude increase in accuracy over conventional methods including the Linear Predictive Method and the Gabor-Volterra-Weiner Polynomial Method. Deterministic chaos is thought to be involved in many physical situations including the onset of turbulence in fluids, chemical reactions and plasma physics. Secondly, we demonstrate the use of the formalism in nonlinear system modelling by providing a graphic example in which it is clear that the neural network has accurately modelled the nonlinear transfer function. It is interesting to note that the formalism provides explicit, analytic, global, approximations to the nonlinear maps underlying the various time series. Furthermore, the neural net seems to be extremely parsimonious in its requirements for data points from the time series. We show that the neural net is able to perform well because it globally approximates the relevant maps by performing a kind of generalized mode decomposition of the maps. 24 refs., 13 figs.

Synchronization of complex delayed dynamical networks with nonlinearly coupled nodes

International Nuclear Information System (INIS)

Liu Tao; Zhao Jun; Hill, David J.

2009-01-01

In this paper, we study the global synchronization of nonlinearly coupled complex delayed dynamical networks with both directed and undirected graphs. Via Lyapunov-Krasovskii stability theory and the network topology, we investigate the global synchronization of such networks. Under the assumption that coupling coefficients are known, a family of delay-independent decentralized nonlinear feedback controllers are designed to globally synchronize the networks. When coupling coefficients are unavailable, an adaptive mechanism is introduced to synthesize a family of delay-independent decentralized adaptive controllers which guarantee the global synchronization of the uncertain networks. Two numerical examples of directed and undirected delayed dynamical network are given, respectively, using the Lorenz system as the nodes of the networks, which demonstrate the effectiveness of proposed results.

Transforming Musical Signals through a Genre Classifying Convolutional Neural Network

Science.gov (United States)

Geng, S.; Ren, G.; Ogihara, M.

2017-05-01

Convolutional neural networks (CNNs) have been successfully applied on both discriminative and generative modeling for music-related tasks. For a particular task, the trained CNN contains information representing the decision making or the abstracting process. One can hope to manipulate existing music based on this 'informed' network and create music with new features corresponding to the knowledge obtained by the network. In this paper, we propose a method to utilize the stored information from a CNN trained on musical genre classification task. The network was composed of three convolutional layers, and was trained to classify five-second song clips into five different genres. After training, randomly selected clips were modified by maximizing the sum of outputs from the network layers. In addition to the potential of such CNNs to produce interesting audio transformation, more information about the network and the original music could be obtained from the analysis of the generated features since these features indicate how the network 'understands' the music.

Nonlinear identification of process dynamics using neural networks

International Nuclear Information System (INIS)

Parlos, A.G.; Atiya, A.F.; Chong, K.T.

1992-01-01

In this paper the nonlinear identification of process dynamics encountered in nuclear power plant components is addressed, in an input-output sense, using artificial neural systems. A hybrid feedforward/feedback neural network, namely, a recurrent multilayer perceptron, is used as the model structure to be identified. The feedforward portion of the network architecture provides its well-known interpolation property, while through recurrency and cross-talk, the local information feedback enables representation of temporal variations in the system nonlinearities. The standard backpropagation learning algorithm is modified, and it is used for the supervised training of the proposed hybrid network. The performance of recurrent multilayer perceptron networks in identifying process dynamics is investigated via the case study of a U-tube steam generator. The response of representative steam generator is predicted using a neural network, and it is compared to the response obtained from a sophisticated computer model based on first principles. The transient responses compare well, although further research is warranted to determine the predictive capabilities of these networks during more severe operational transients and accident scenarios

Nonlinear noninteger order circuits and systems an introduction

CERN Document Server

Arena, P; Fortuna, L; Porto, D

2001-01-01

In this book, the reader will find a theoretical introduction to noninteger order systems, as well as several applications showing their features and peculiarities. The main definitions and results of research on noninteger order systems and modelling of physical noninteger phenomena are reported together with problems of their approximation. Control applications, noninteger order CNNs and circuit realizations of noninteger order systems are also presented.The book is intended for students and researchers involved in the simulation and control of nonlinear noninteger order systems, with partic

Artificial Neural Networks for Nonlinear Dynamic Response Simulation in Mechanical Systems

DEFF Research Database (Denmark)

Christiansen, Niels Hørbye; Høgsberg, Jan Becker; Winther, Ole

2011-01-01

It is shown how artificial neural networks can be trained to predict dynamic response of a simple nonlinear structure. Data generated using a nonlinear finite element model of a simplified wind turbine is used to train a one layer artificial neural network. When trained properly the network is ab...... to perform accurate response prediction much faster than the corresponding finite element model. Initial result indicate a reduction in cpu time by two orders of magnitude....

Implementation of neural network based non-linear predictive

DEFF Research Database (Denmark)

Sørensen, Paul Haase; Nørgård, Peter Magnus; Ravn, Ole

1998-01-01

The paper describes a control method for non-linear systems based on generalized predictive control. Generalized predictive control (GPC) was developed to control linear systems including open loop unstable and non-minimum phase systems, but has also been proposed extended for the control of non......-linear systems. GPC is model-based and in this paper we propose the use of a neural network for the modeling of the system. Based on the neural network model a controller with extended control horizon is developed and the implementation issues are discussed, with particular emphasis on an efficient Quasi......-Newton optimization algorithm. The performance is demonstrated on a pneumatic servo system....

Extended LaSalle's Invariance Principle for Full-Range Cellular Neural Networks

Directory of Open Access Journals (Sweden)

Mauro Di Marco

2009-01-01

Full Text Available In several relevant applications to the solution of signal processing tasks in real time, a cellular neural network (CNN is required to be convergent, that is, each solution should tend toward some equilibrium point. The paper develops a Lyapunov method, which is based on a generalized version of LaSalle's invariance principle, for studying convergence and stability of the differential inclusions modeling the dynamics of the full-range (FR model of CNNs. The applicability of the method is demonstrated by obtaining a rigorous proof of convergence for symmetric FR-CNNs. The proof, which is a direct consequence of the fact that a symmetric FR-CNN admits a strict Lyapunov function, is much more simple than the corresponding proof of convergence for symmetric standard CNNs.

Understanding the Convolutional Neural Networks with Gradient Descent and Backpropagation

Science.gov (United States)

Zhou, XueFei

2018-04-01

With the development of computer technology, the applications of machine learning are more and more extensive. And machine learning is providing endless opportunities to develop new applications. One of those applications is image recognition by using Convolutional Neural Networks (CNNs). CNN is one of the most common algorithms in image recognition. It is significant to understand its theory and structure for every scholar who is interested in this field. CNN is mainly used in computer identification, especially in voice, text recognition and other aspects of the application. It utilizes hierarchical structure with different layers to accelerate computing speed. In addition, the greatest features of CNNs are the weight sharing and dimension reduction. And all of these consolidate the high effectiveness and efficiency of CNNs with idea computing speed and error rate. With the help of other learning altruisms, CNNs could be used in several scenarios for machine learning, especially for deep learning. Based on the general introduction to the background and the core solution CNN, this paper is going to focus on summarizing how Gradient Descent and Backpropagation work, and how they contribute to the high performances of CNNs. Also, some practical applications will be discussed in the following parts. The last section exhibits the conclusion and some perspectives of future work.

Training Deep Convolutional Neural Networks with Resistive Cross-Point Devices.

Science.gov (United States)

Gokmen, Tayfun; Onen, Murat; Haensch, Wilfried

2017-01-01

In a previous work we have detailed the requirements for obtaining maximal deep learning performance benefit by implementing fully connected deep neural networks (DNN) in the form of arrays of resistive devices. Here we extend the concept of Resistive Processing Unit (RPU) devices to convolutional neural networks (CNNs). We show how to map the convolutional layers to fully connected RPU arrays such that the parallelism of the hardware can be fully utilized in all three cycles of the backpropagation algorithm. We find that the noise and bound limitations imposed by the analog nature of the computations performed on the arrays significantly affect the training accuracy of the CNNs. Noise and bound management techniques are presented that mitigate these problems without introducing any additional complexity in the analog circuits and that can be addressed by the digital circuits. In addition, we discuss digitally programmable update management and device variability reduction techniques that can be used selectively for some of the layers in a CNN. We show that a combination of all those techniques enables a successful application of the RPU concept for training CNNs. The techniques discussed here are more general and can be applied beyond CNN architectures and therefore enables applicability of the RPU approach to a large class of neural network architectures.

Training Deep Convolutional Neural Networks with Resistive Cross-Point Devices

Science.gov (United States)

Gokmen, Tayfun; Onen, Murat; Haensch, Wilfried

2017-01-01

In a previous work we have detailed the requirements for obtaining maximal deep learning performance benefit by implementing fully connected deep neural networks (DNN) in the form of arrays of resistive devices. Here we extend the concept of Resistive Processing Unit (RPU) devices to convolutional neural networks (CNNs). We show how to map the convolutional layers to fully connected RPU arrays such that the parallelism of the hardware can be fully utilized in all three cycles of the backpropagation algorithm. We find that the noise and bound limitations imposed by the analog nature of the computations performed on the arrays significantly affect the training accuracy of the CNNs. Noise and bound management techniques are presented that mitigate these problems without introducing any additional complexity in the analog circuits and that can be addressed by the digital circuits. In addition, we discuss digitally programmable update management and device variability reduction techniques that can be used selectively for some of the layers in a CNN. We show that a combination of all those techniques enables a successful application of the RPU concept for training CNNs. The techniques discussed here are more general and can be applied beyond CNN architectures and therefore enables applicability of the RPU approach to a large class of neural network architectures. PMID:29066942

Networked Predictive Control for Nonlinear Systems With Arbitrary Region Quantizers.

Science.gov (United States)

Yang, Hongjiu; Xu, Yang; Xia, Yuanqing; Zhang, Jinhui

2017-04-06

In this paper, networked predictive control is investigated for planar nonlinear systems with quantization by an extended state observer (ESO). The ESO is used not only to deal with nonlinear terms but also to generate predictive states for dealing with network-induced delays. Two arbitrary region quantizers are applied to take effective values of signals in forward channel and feedback channel, respectively. Based on a "zoom" strategy, sufficient conditions are given to guarantee stabilization of the closed-loop networked control system with quantization. A simulation example is proposed to exhibit advantages and availability of the results.

Quasiperiodic AlGaAs superlattices for neuromorphic networks and nonlinear control systems

Energy Technology Data Exchange (ETDEWEB)

Malyshev, K. V., E-mail: malyshev@bmstu.ru [Electronics and Laser Technology Department, Bauman Moscow State Technical University, Moscow 105005 (Russian Federation)

2015-01-28

The application of quasiperiodic AlGaAs superlattices as a nonlinear element of the FitzHugh–Nagumo neuromorphic network is proposed and theoretically investigated on the example of Fibonacci and figurate superlattices. The sequences of symbols for the figurate superlattices were produced by decomposition of the Fibonacci superlattices' symbolic sequences. A length of each segment of the decomposition was equal to the corresponding figurate number. It is shown that a nonlinear network based upon Fibonacci and figurate superlattices provides better parallel filtration of a half-tone picture; then, a network based upon traditional diodes which have cubic voltage-current characteristics. It was found that the figurate superlattice F{sup 0}{sub 11}(1) as a nonlinear network's element provides the filtration error almost twice less than the conventional “cubic” diode. These advantages are explained by a wavelike shape of the decreasing part of the quasiperiodic superlattice's voltage-current characteristic, which leads to multistability of the network's cell. This multistability promises new interesting nonlinear dynamical phenomena. A variety of wavy forms of voltage-current characteristics opens up new interesting possibilities for quasiperiodic superlattices and especially for figurate superlattices in many areas—from nervous system modeling to nonlinear control systems development.

Feedforward Nonlinear Control Using Neural Gas Network

OpenAIRE

Machón-González, Iván; López-García, Hilario

2017-01-01

Nonlinear systems control is a main issue in control theory. Many developed applications suffer from a mathematical foundation not as general as the theory of linear systems. This paper proposes a control strategy of nonlinear systems with unknown dynamics by means of a set of local linear models obtained by a supervised neural gas network. The proposed approach takes advantage of the neural gas feature by which the algorithm yields a very robust clustering procedure. The direct model of the ...

Deep Convolutional Networks for Event Reconstruction and Particle Tagging on NOvA and DUNE

CERN Multimedia

CERN. Geneva

2017-01-01

Deep Convolutional Neural Networks (CNNs) have been widely applied in computer vision to solve complex problems in image recognition and analysis. In recent years many efforts have emerged to extend the use of this technology to HEP applications, including the Convolutional Visual Network (CVN), our implementation for identification of neutrino events. In this presentation I will describe the core concepts of CNNs, the details of our particular implementation in the Caffe framework and our application to identify NOvA events. NOvA is a long baseline neutrino experiment whose main goal is the measurement of neutrino oscillations. This relies on the accurate identification and reconstruction of the neutrino flavor in the interactions we observe. In 2016 the NOvA experiment released results for the observation of oscillations in the ν μ → ν e channel, the first HEP result employing CNNs. I will also discuss our approach at event identification on NOvA as well as recent developments in the application of CNN...

Chinese Medical Question Answer Matching Using End-to-End Character-Level Multi-Scale CNNs

Directory of Open Access Journals (Sweden)

Sheng Zhang

2017-07-01

Full Text Available This paper focuses mainly on the problem of Chinese medical question answer matching, which is arguably more challenging than open-domain question answer matching in English due to the combination of its domain-restricted nature and the language-specific features of Chinese. We present an end-to-end character-level multi-scale convolutional neural framework in which character embeddings instead of word embeddings are used to avoid Chinese word segmentation in text preprocessing, and multi-scale convolutional neural networks (CNNs are then introduced to extract contextual information from either question or answer sentences over different scales. The proposed framework can be trained with minimal human supervision and does not require any handcrafted features, rule-based patterns, or external resources. To validate our framework, we create a new text corpus, named cMedQA, by harvesting questions and answers from an online Chinese health and wellness community. The experimental results on the cMedQA dataset show that our framework significantly outperforms several strong baselines, and achieves an improvement of top-1 accuracy by up to 19%.

Conversion of Continuous-Valued Deep Networks to Efficient Event-Driven Networks for Image Classification.

Science.gov (United States)

Rueckauer, Bodo; Lungu, Iulia-Alexandra; Hu, Yuhuang; Pfeiffer, Michael; Liu, Shih-Chii

2017-01-01

Spiking neural networks (SNNs) can potentially offer an efficient way of doing inference because the neurons in the networks are sparsely activated and computations are event-driven. Previous work showed that simple continuous-valued deep Convolutional Neural Networks (CNNs) can be converted into accurate spiking equivalents. These networks did not include certain common operations such as max-pooling, softmax, batch-normalization and Inception-modules. This paper presents spiking equivalents of these operations therefore allowing conversion of nearly arbitrary CNN architectures. We show conversion of popular CNN architectures, including VGG-16 and Inception-v3, into SNNs that produce the best results reported to date on MNIST, CIFAR-10 and the challenging ImageNet dataset. SNNs can trade off classification error rate against the number of available operations whereas deep continuous-valued neural networks require a fixed number of operations to achieve their classification error rate. From the examples of LeNet for MNIST and BinaryNet for CIFAR-10, we show that with an increase in error rate of a few percentage points, the SNNs can achieve more than 2x reductions in operations compared to the original CNNs. This highlights the potential of SNNs in particular when deployed on power-efficient neuromorphic spiking neuron chips, for use in embedded applications.

Conversion of Continuous-Valued Deep Networks to Efficient Event-Driven Networks for Image Classification

Directory of Open Access Journals (Sweden)

Bodo Rueckauer

2017-12-01

Full Text Available Spiking neural networks (SNNs can potentially offer an efficient way of doing inference because the neurons in the networks are sparsely activated and computations are event-driven. Previous work showed that simple continuous-valued deep Convolutional Neural Networks (CNNs can be converted into accurate spiking equivalents. These networks did not include certain common operations such as max-pooling, softmax, batch-normalization and Inception-modules. This paper presents spiking equivalents of these operations therefore allowing conversion of nearly arbitrary CNN architectures. We show conversion of popular CNN architectures, including VGG-16 and Inception-v3, into SNNs that produce the best results reported to date on MNIST, CIFAR-10 and the challenging ImageNet dataset. SNNs can trade off classification error rate against the number of available operations whereas deep continuous-valued neural networks require a fixed number of operations to achieve their classification error rate. From the examples of LeNet for MNIST and BinaryNet for CIFAR-10, we show that with an increase in error rate of a few percentage points, the SNNs can achieve more than 2x reductions in operations compared to the original CNNs. This highlights the potential of SNNs in particular when deployed on power-efficient neuromorphic spiking neuron chips, for use in embedded applications.

Linear and nonlinear ARMA model parameter estimation using an artificial neural network

Science.gov (United States)

Chon, K. H.; Cohen, R. J.

1997-01-01

This paper addresses parametric system identification of linear and nonlinear dynamic systems by analysis of the input and output signals. Specifically, we investigate the relationship between estimation of the system using a feedforward neural network model and estimation of the system by use of linear and nonlinear autoregressive moving-average (ARMA) models. By utilizing a neural network model incorporating a polynomial activation function, we show the equivalence of the artificial neural network to the linear and nonlinear ARMA models. We compare the parameterization of the estimated system using the neural network and ARMA approaches by utilizing data generated by means of computer simulations. Specifically, we show that the parameters of a simulated ARMA system can be obtained from the neural network analysis of the simulated data or by conventional least squares ARMA analysis. The feasibility of applying neural networks with polynomial activation functions to the analysis of experimental data is explored by application to measurements of heart rate (HR) and instantaneous lung volume (ILV) fluctuations.

Lukasiewicz-Topos Models of Neural Networks, Cell Genome and Interactome Nonlinear Dynamic Models

CERN Document Server

Baianu, I C

2004-01-01

A categorical and Lukasiewicz-Topos framework for Lukasiewicz Algebraic Logic models of nonlinear dynamics in complex functional systems such as neural networks, genomes and cell interactomes is proposed. Lukasiewicz Algebraic Logic models of genetic networks and signaling pathways in cells are formulated in terms of nonlinear dynamic systems with n-state components that allow for the generalization of previous logical models of both genetic activities and neural networks. An algebraic formulation of variable 'next-state functions' is extended to a Lukasiewicz Topos with an n-valued Lukasiewicz Algebraic Logic subobject classifier description that represents non-random and nonlinear network activities as well as their transformations in developmental processes and carcinogenesis.

Bilinear Convolutional Neural Networks for Fine-grained Visual Recognition.

Science.gov (United States)

Lin, Tsung-Yu; RoyChowdhury, Aruni; Maji, Subhransu

2017-07-04

We present a simple and effective architecture for fine-grained recognition called Bilinear Convolutional Neural Networks (B-CNNs). These networks represent an image as a pooled outer product of features derived from two CNNs and capture localized feature interactions in a translationally invariant manner. B-CNNs are related to orderless texture representations built on deep features but can be trained in an end-to-end manner. Our most accurate model obtains 84.1%, 79.4%, 84.5% and 91.3% per-image accuracy on the Caltech-UCSD birds [66], NABirds [63], FGVC aircraft [42], and Stanford cars [33] dataset respectively and runs at 30 frames-per-second on a NVIDIA Titan X GPU. We then present a systematic analysis of these networks and show that (1) the bilinear features are highly redundant and can be reduced by an order of magnitude in size without significant loss in accuracy, (2) are also effective for other image classification tasks such as texture and scene recognition, and (3) can be trained from scratch on the ImageNet dataset offering consistent improvements over the baseline architecture. Finally, we present visualizations of these models on various datasets using top activations of neural units and gradient-based inversion techniques. The source code for the complete system is available at http://vis-www.cs.umass.edu/bcnn.

Image processing with cellular nonlinear networks implemented on field-programmable gate arrays for real-time applications in nuclear fusion

International Nuclear Information System (INIS)

Palazzo, S.; Vagliasindi, G.; Arena, P.; Murari, A.; Mazon, D.; De Maack, A.

2010-01-01

In the past years cameras have become increasingly common tools in scientific applications. They are now quite systematically used in magnetic confinement fusion, to the point that infrared imaging is starting to be used systematically for real-time machine protection in major devices. However, in order to guarantee that the control system can always react rapidly in case of critical situations, the time required for the processing of the images must be as predictable as possible. The approach described in this paper combines the new computational paradigm of cellular nonlinear networks (CNNs) with field-programmable gate arrays and has been tested in an application for the detection of hot spots on the plasma facing components in JET. The developed system is able to perform real-time hot spot recognition, by processing the image stream captured by JET wide angle infrared camera, with the guarantee that computational time is constant and deterministic. The statistical results obtained from a quite extensive set of examples show that this solution approximates very well an ad hoc serial software algorithm, with no false or missed alarms and an almost perfect overlapping of alarm intervals. The computational time can be reduced to a millisecond time scale for 8 bit 496x560-sized images. Moreover, in our implementation, the computational time, besides being deterministic, is practically independent of the number of iterations performed by the CNN - unlike software CNN implementations.

Neonatal Seizure Detection Using Deep Convolutional Neural Networks.

Science.gov (United States)

Ansari, Amir H; Cherian, Perumpillichira J; Caicedo, Alexander; Naulaers, Gunnar; De Vos, Maarten; Van Huffel, Sabine

2018-04-02

Identifying a core set of features is one of the most important steps in the development of an automated seizure detector. In most of the published studies describing features and seizure classifiers, the features were hand-engineered, which may not be optimal. The main goal of the present paper is using deep convolutional neural networks (CNNs) and random forest to automatically optimize feature selection and classification. The input of the proposed classifier is raw multi-channel EEG and the output is the class label: seizure/nonseizure. By training this network, the required features are optimized, while fitting a nonlinear classifier on the features. After training the network with EEG recordings of 26 neonates, five end layers performing the classification were replaced with a random forest classifier in order to improve the performance. This resulted in a false alarm rate of 0.9 per hour and seizure detection rate of 77% using a test set of EEG recordings of 22 neonates that also included dubious seizures. The newly proposed CNN classifier outperformed three data-driven feature-based approaches and performed similar to a previously developed heuristic method.

Implementation of neural network based non-linear predictive control

DEFF Research Database (Denmark)

Sørensen, Paul Haase; Nørgård, Peter Magnus; Ravn, Ole

1999-01-01

This paper describes a control method for non-linear systems based on generalized predictive control. Generalized predictive control (GPC) was developed to control linear systems, including open-loop unstable and non-minimum phase systems, but has also been proposed to be extended for the control...... of non-linear systems. GPC is model based and in this paper we propose the use of a neural network for the modeling of the system. Based on the neural network model, a controller with extended control horizon is developed and the implementation issues are discussed, with particular emphasis...... on an efficient quasi-Newton algorithm. The performance is demonstrated on a pneumatic servo system....

Non-Linear State Estimation Using Pre-Trained Neural Networks

DEFF Research Database (Denmark)

Bayramoglu, Enis; Andersen, Nils Axel; Ravn, Ole

2010-01-01

effecting the transformation. This function is approximated by a neural network using offline training. The training is based on monte carlo sampling. A way to obtain parametric distributions of flexible shape to be used easily with these networks is also presented. The method can also be used to improve...... other parametric methods around regions with strong non-linearities by including them inside the network....

Training Deep Convolutional Neural Networks with Resistive Cross-Point Devices

Directory of Open Access Journals (Sweden)

Tayfun Gokmen

2017-10-01

Full Text Available In a previous work we have detailed the requirements for obtaining maximal deep learning performance benefit by implementing fully connected deep neural networks (DNN in the form of arrays of resistive devices. Here we extend the concept of Resistive Processing Unit (RPU devices to convolutional neural networks (CNNs. We show how to map the convolutional layers to fully connected RPU arrays such that the parallelism of the hardware can be fully utilized in all three cycles of the backpropagation algorithm. We find that the noise and bound limitations imposed by the analog nature of the computations performed on the arrays significantly affect the training accuracy of the CNNs. Noise and bound management techniques are presented that mitigate these problems without introducing any additional complexity in the analog circuits and that can be addressed by the digital circuits. In addition, we discuss digitally programmable update management and device variability reduction techniques that can be used selectively for some of the layers in a CNN. We show that a combination of all those techniques enables a successful application of the RPU concept for training CNNs. The techniques discussed here are more general and can be applied beyond CNN architectures and therefore enables applicability of the RPU approach to a large class of neural network architectures.

Deciphering the imprint of topology on nonlinear dynamical network stability

International Nuclear Information System (INIS)

Nitzbon, J; Schultz, P; Heitzig, J; Kurths, J; Hellmann, F

2017-01-01

Coupled oscillator networks show complex interrelations between topological characteristics of the network and the nonlinear stability of single nodes with respect to large but realistic perturbations. We extend previous results on these relations by incorporating sampling-based measures of the transient behaviour of the system, its survivability, as well as its asymptotic behaviour, its basin stability. By combining basin stability and survivability we uncover novel, previously unknown asymptotic states with solitary, desynchronized oscillators which are rotating with a frequency different from their natural one. They occur almost exclusively after perturbations at nodes with specific topological properties. More generally we confirm and significantly refine the results on the distinguished role tree-shaped appendices play for nonlinear stability. We find a topological classification scheme for nodes located in such appendices, that exactly separates them according to their stability properties, thus establishing a strong link between topology and dynamics. Hence, the results can be used for the identification of vulnerable nodes in power grids or other coupled oscillator networks. From this classification we can derive general design principles for resilient power grids. We find that striving for homogeneous network topologies facilitates a better performance in terms of nonlinear dynamical network stability. While the employed second-order Kuramoto-like model is parametrised to be representative for power grids, we expect these insights to transfer to other critical infrastructure systems or complex network dynamics appearing in various other fields. (paper)

Nonlinear recurrent neural networks for finite-time solution of general time-varying linear matrix equations.

Science.gov (United States)

Xiao, Lin; Liao, Bolin; Li, Shuai; Chen, Ke

2018-02-01

In order to solve general time-varying linear matrix equations (LMEs) more efficiently, this paper proposes two nonlinear recurrent neural networks based on two nonlinear activation functions. According to Lyapunov theory, such two nonlinear recurrent neural networks are proved to be convergent within finite-time. Besides, by solving differential equation, the upper bounds of the finite convergence time are determined analytically. Compared with existing recurrent neural networks, the proposed two nonlinear recurrent neural networks have a better convergence property (i.e., the upper bound is lower), and thus the accurate solutions of general time-varying LMEs can be obtained with less time. At last, various different situations have been considered by setting different coefficient matrices of general time-varying LMEs and a great variety of computer simulations (including the application to robot manipulators) have been conducted to validate the better finite-time convergence of the proposed two nonlinear recurrent neural networks. Copyright © 2017 Elsevier Ltd. All rights reserved.

Synchronization in Complex Networks of Nonlinear Dynamical Systems

CERN Document Server

Wu, Chai Wah

2007-01-01

This book brings together two emerging research areas: synchronization in coupled nonlinear systems and complex networks, and study conditions under which a complex network of dynamical systems synchronizes. While there are many texts that study synchronization in chaotic systems or properties of complex networks, there are few texts that consider the intersection of these two very active and interdisciplinary research areas. The main theme of this book is that synchronization conditions can be related to graph theoretical properties of the underlying coupling topology. The book introduces ide

THEORETICAL EVALUATION OF NONLINEAR EFFECTS ON OPTICAL WDM NETWORKS WITH VARIOUS FIBER TYPES

Directory of Open Access Journals (Sweden)

YASIN M. KARFAA

2010-09-01

Full Text Available A theoretical study is carried out to evaluate the performance of an opticalwavelength division multiplexing (WDM network transmission system in the presenceof crosstalk due to optical fiber nonlinearities. The most significant nonlinear effects inthe optical fiber which are Cross-Phase Modulation (XPM, Four-Wave Mixing (FWM,and Stimulated Raman Scattering (SRS are investigated. Four types of optical fiber areincluded in the analysis; these are: single-mode fiber (SMF, dispersion compensationfiber (DCF, non-zero dispersion fiber (NZDF, and non-zero dispersion shifted fiber(NZDSF. The results represent the standard deviation of nonlinearity induced crosstalknoise power due to FWM and SRS, XPM power penalty for SMF, DCF, NZDF, andNZDSF types of fiber, besides the Bit Error Rate (BER for the three nonlinear effectsusing standard fiber type (SMF. It is concluded that three significant fiber nonlinearitiesare making huge limitations against increasing the launched power which is desired,otherwise, lower values of launched power limit network expansion including length,distance, covered areas, and number of users accessing the WDM network, unlesssuitable precautions are taken to neutralize the nonlinear effects. Besides, various fibertypes are not behaving similarly towards network parameters.

Cerebral vessels segmentation for light-sheet microscopy image using convolutional neural networks

Science.gov (United States)

Hu, Chaoen; Hui, Hui; Wang, Shuo; Dong, Di; Liu, Xia; Yang, Xin; Tian, Jie

2017-03-01

Cerebral vessel segmentation is an important step in image analysis for brain function and brain disease studies. To extract all the cerebrovascular patterns, including arteries and capillaries, some filter-based methods are used to segment vessels. However, the design of accurate and robust vessel segmentation algorithms is still challenging, due to the variety and complexity of images, especially in cerebral blood vessel segmentation. In this work, we addressed a problem of automatic and robust segmentation of cerebral micro-vessels structures in cerebrovascular images acquired by light-sheet microscope for mouse. To segment micro-vessels in large-scale image data, we proposed a convolutional neural networks (CNNs) architecture trained by 1.58 million pixels with manual label. Three convolutional layers and one fully connected layer were used in the CNNs model. We extracted a patch of size 32x32 pixels in each acquired brain vessel image as training data set to feed into CNNs for classification. This network was trained to output the probability that the center pixel of input patch belongs to vessel structures. To build the CNNs architecture, a series of mouse brain vascular images acquired from a commercial light sheet fluorescence microscopy (LSFM) system were used for training the model. The experimental results demonstrated that our approach is a promising method for effectively segmenting micro-vessels structures in cerebrovascular images with vessel-dense, nonuniform gray-level and long-scale contrast regions.

Nonlinear stochastic systems with network-induced phenomena recursive filtering and sliding-mode design

CERN Document Server

Hu, Jun; Gao, Huijun

2014-01-01

This monograph introduces methods for handling filtering and control problems in nonlinear stochastic systems arising from network-induced phenomena consequent on limited communication capacity. Such phenomena include communication delay, packet dropout, signal quantization or saturation, randomly occurring nonlinearities and randomly occurring uncertainties.The text is self-contained, beginning with an introduction to nonlinear stochastic systems, network-induced phenomena and filtering and control, moving through a collection of the latest research results which focuses on the three aspects

Noise Response Data Reveal Novel Controllability Gramian for Nonlinear Network Dynamics

Science.gov (United States)

Kashima, Kenji

2016-01-01

Control of nonlinear large-scale dynamical networks, e.g., collective behavior of agents interacting via a scale-free connection topology, is a central problem in many scientific and engineering fields. For the linear version of this problem, the so-called controllability Gramian has played an important role to quantify how effectively the dynamical states are reachable by a suitable driving input. In this paper, we first extend the notion of the controllability Gramian to nonlinear dynamics in terms of the Gibbs distribution. Next, we show that, when the networks are open to environmental noise, the newly defined Gramian is equal to the covariance matrix associated with randomly excited, but uncontrolled, dynamical state trajectories. This fact theoretically justifies a simple Monte Carlo simulation that can extract effectively controllable subdynamics in nonlinear complex networks. In addition, the result provides a novel insight into the relationship between controllability and statistical mechanics. PMID:27264780

A locality aware convolutional neural networks accelerator

NARCIS (Netherlands)

Shi, R.; Xu, Z.; Sun, Z.; Peemen, M.C.J.; Li, A.; Corporaal, H.; Wu, D.

2015-01-01

The advantages of Convolutional Neural Networks (CNNs) with respect to traditional methods for visual pattern recognition have changed the field of machine vision. The main issue that hinders broad adoption of this technique is the massive computing workload in CNN that prevents real-time

Feedforward Nonlinear Control Using Neural Gas Network

Directory of Open Access Journals (Sweden)

Iván Machón-González

2017-01-01

Full Text Available Nonlinear systems control is a main issue in control theory. Many developed applications suffer from a mathematical foundation not as general as the theory of linear systems. This paper proposes a control strategy of nonlinear systems with unknown dynamics by means of a set of local linear models obtained by a supervised neural gas network. The proposed approach takes advantage of the neural gas feature by which the algorithm yields a very robust clustering procedure. The direct model of the plant constitutes a piece-wise linear approximation of the nonlinear system and each neuron represents a local linear model for which a linear controller is designed. The neural gas model works as an observer and a controller at the same time. A state feedback control is implemented by estimation of the state variables based on the local transfer function that was provided by the local linear model. The gradient vectors obtained by the supervised neural gas algorithm provide a robust procedure for feedforward nonlinear control, that is, supposing the inexistence of disturbances.

FEM-based neural-network approach to nonlinear modeling with application to longitudinal vehicle dynamics control.

Science.gov (United States)

Kalkkuhl, J; Hunt, K J; Fritz, H

1999-01-01

An finite-element methods (FEM)-based neural-network approach to Nonlinear AutoRegressive with eXogenous input (NARX) modeling is presented. The method uses multilinear interpolation functions on C0 rectangular elements. The local and global structure of the resulting model is analyzed. It is shown that the model can be interpreted both as a local model network and a single layer feedforward neural network. The main aim is to use the model for nonlinear control design. The proposed FEM NARX description is easily accessible to feedback linearizing control techniques. Its use with a two-degrees of freedom nonlinear internal model controller is discussed. The approach is applied to modeling of the nonlinear longitudinal dynamics of an experimental lorry, using measured data. The modeling results are compared with local model network and multilayer perceptron approaches. A nonlinear speed controller was designed based on the identified FEM model. The controller was implemented in a test vehicle, and several experimental results are presented.

Distributed Synchronization in Networks of Agent Systems With Nonlinearities and Random Switchings.

Science.gov (United States)

Tang, Yang; Gao, Huijun; Zou, Wei; Kurths, Jürgen

2013-02-01

In this paper, the distributed synchronization problem of networks of agent systems with controllers and nonlinearities subject to Bernoulli switchings is investigated. Controllers and adaptive updating laws injected in each vertex of networks depend on the state information of its neighborhood. Three sets of Bernoulli stochastic variables are introduced to describe the occurrence probabilities of distributed adaptive controllers, updating laws and nonlinearities, respectively. By the Lyapunov functions method, we show that the distributed synchronization of networks composed of agent systems with multiple randomly occurring nonlinearities, multiple randomly occurring controllers, and multiple randomly occurring updating laws can be achieved in mean square under certain criteria. The conditions derived in this paper can be solved by semi-definite programming. Moreover, by mathematical analysis, we find that the coupling strength, the probabilities of the Bernoulli stochastic variables, and the form of nonlinearities have great impacts on the convergence speed and the terminal control strength. The synchronization criteria and the observed phenomena are demonstrated by several numerical simulation examples. In addition, the advantage of distributed adaptive controllers over conventional adaptive controllers is illustrated.

Bio-inspired spiking neural network for nonlinear systems control.

Science.gov (United States)

Pérez, Javier; Cabrera, Juan A; Castillo, Juan J; Velasco, Juan M

2018-08-01

Spiking neural networks (SNN) are the third generation of artificial neural networks. SNN are the closest approximation to biological neural networks. SNNs make use of temporal spike trains to command inputs and outputs, allowing a faster and more complex computation. As demonstrated by biological organisms, they are a potentially good approach to designing controllers for highly nonlinear dynamic systems in which the performance of controllers developed by conventional techniques is not satisfactory or difficult to implement. SNN-based controllers exploit their ability for online learning and self-adaptation to evolve when transferred from simulations to the real world. SNN's inherent binary and temporary way of information codification facilitates their hardware implementation compared to analog neurons. Biological neural networks often require a lower number of neurons compared to other controllers based on artificial neural networks. In this work, these neuronal systems are imitated to perform the control of non-linear dynamic systems. For this purpose, a control structure based on spiking neural networks has been designed. Particular attention has been paid to optimizing the structure and size of the neural network. The proposed structure is able to control dynamic systems with a reduced number of neurons and connections. A supervised learning process using evolutionary algorithms has been carried out to perform controller training. The efficiency of the proposed network has been verified in two examples of dynamic systems control. Simulations show that the proposed control based on SNN exhibits superior performance compared to other approaches based on Neural Networks and SNNs. Copyright © 2018 Elsevier Ltd. All rights reserved.

Model-free inference of direct network interactions from nonlinear collective dynamics.

Science.gov (United States)

Casadiego, Jose; Nitzan, Mor; Hallerberg, Sarah; Timme, Marc

2017-12-19

The topology of interactions in network dynamical systems fundamentally underlies their function. Accelerating technological progress creates massively available data about collective nonlinear dynamics in physical, biological, and technological systems. Detecting direct interaction patterns from those dynamics still constitutes a major open problem. In particular, current nonlinear dynamics approaches mostly require to know a priori a model of the (often high dimensional) system dynamics. Here we develop a model-independent framework for inferring direct interactions solely from recording the nonlinear collective dynamics generated. Introducing an explicit dependency matrix in combination with a block-orthogonal regression algorithm, the approach works reliably across many dynamical regimes, including transient dynamics toward steady states, periodic and non-periodic dynamics, and chaos. Together with its capabilities to reveal network (two point) as well as hypernetwork (e.g., three point) interactions, this framework may thus open up nonlinear dynamics options of inferring direct interaction patterns across systems where no model is known.

Effects of network structure on the synchronizability of nonlinearly coupled Hindmarsh–Rose neurons

International Nuclear Information System (INIS)

Li, Chun-Hsien; Yang, Suh-Yuh

2015-01-01

This work is devoted to investigate the effects of network structure on the synchronizability of nonlinearly coupled dynamical network of Hindmarsh–Rose neurons with a sigmoidal coupling function. We mainly focus on the networks that exhibit the small-world character or scale-free property. By checking the first nonzero eigenvalue of the outer-coupling matrix, which is closely related to the synchronization threshold, the synchronizabilities of three specific network ensembles with prescribed network structures are compared. Interestingly, we find that networks with more connections will not necessarily result in better synchronizability. - Highlights: • We investigate the effects of network structure on the synchronizability of nonlinearly coupled Hindmarsh–Rose neurons. • We mainly consider the networks that exhibit the small-world character or scale-free property. • The synchronizability of three specific network ensembles with prescribed network structures are compared. • Networks with more connections will not necessarily result in better synchronizability

Effects of network structure on the synchronizability of nonlinearly coupled Hindmarsh–Rose neurons

Energy Technology Data Exchange (ETDEWEB)

Li, Chun-Hsien, E-mail: chli@nknucc.nknu.edu.tw [Department of Mathematics, National Kaohsiung Normal University, Yanchao District, Kaohsiung City 82444, Taiwan (China); Yang, Suh-Yuh, E-mail: syyang@math.ncu.edu.tw [Department of Mathematics, National Central University, Jhongli District, Taoyuan City 32001, Taiwan (China)

2015-10-23

This work is devoted to investigate the effects of network structure on the synchronizability of nonlinearly coupled dynamical network of Hindmarsh–Rose neurons with a sigmoidal coupling function. We mainly focus on the networks that exhibit the small-world character or scale-free property. By checking the first nonzero eigenvalue of the outer-coupling matrix, which is closely related to the synchronization threshold, the synchronizabilities of three specific network ensembles with prescribed network structures are compared. Interestingly, we find that networks with more connections will not necessarily result in better synchronizability. - Highlights: • We investigate the effects of network structure on the synchronizability of nonlinearly coupled Hindmarsh–Rose neurons. • We mainly consider the networks that exhibit the small-world character or scale-free property. • The synchronizability of three specific network ensembles with prescribed network structures are compared. • Networks with more connections will not necessarily result in better synchronizability.

A novel nonlinear adaptive filter using a pipelined second-order Volterra recurrent neural network.

Science.gov (United States)

Zhao, Haiquan; Zhang, Jiashu

2009-12-01

To enhance the performance and overcome the heavy computational complexity of recurrent neural networks (RNN), a novel nonlinear adaptive filter based on a pipelined second-order Volterra recurrent neural network (PSOVRNN) is proposed in this paper. A modified real-time recurrent learning (RTRL) algorithm of the proposed filter is derived in much more detail. The PSOVRNN comprises of a number of simple small-scale second-order Volterra recurrent neural network (SOVRNN) modules. In contrast to the standard RNN, these modules of a PSOVRNN can be performed simultaneously in a pipelined parallelism fashion, which can lead to a significant improvement in its total computational efficiency. Moreover, since each module of the PSOVRNN is a SOVRNN in which nonlinearity is introduced by the recursive second-order Volterra (RSOV) expansion, its performance can be further improved. Computer simulations have demonstrated that the PSOVRNN performs better than the pipelined recurrent neural network (PRNN) and RNN for nonlinear colored signals prediction and nonlinear channel equalization. However, the superiority of the PSOVRNN over the PRNN is at the cost of increasing computational complexity due to the introduced nonlinear expansion of each module.

Convolutional Neural Networks for SAR Image Segmentation

DEFF Research Database (Denmark)

Malmgren-Hansen, David; Nobel-Jørgensen, Morten

2015-01-01

Segmentation of Synthetic Aperture Radar (SAR) images has several uses, but it is a difficult task due to a number of properties related to SAR images. In this article we show how Convolutional Neural Networks (CNNs) can easily be trained for SAR image segmentation with good results. Besides...

Information mining in weighted complex networks with nonlinear rating projection

Science.gov (United States)

Liao, Hao; Zeng, An; Zhou, Mingyang; Mao, Rui; Wang, Bing-Hong

2017-10-01

Weighted rating networks are commonly used by e-commerce providers nowadays. In order to generate an objective ranking of online items' quality according to users' ratings, many sophisticated algorithms have been proposed in the complex networks domain. In this paper, instead of proposing new algorithms we focus on a more fundamental problem: the nonlinear rating projection. The basic idea is that even though the rating values given by users are linearly separated, the real preference of users to items between the different given values is nonlinear. We thus design an approach to project the original ratings of users to more representative values. This approach can be regarded as a data pretreatment method. Simulation in both artificial and real networks shows that the performance of the ranking algorithms can be improved when the projected ratings are used.

Nonlinear Model Predictive Control Based on a Self-Organizing Recurrent Neural Network.

Science.gov (United States)

Han, Hong-Gui; Zhang, Lu; Hou, Ying; Qiao, Jun-Fei

2016-02-01

A nonlinear model predictive control (NMPC) scheme is developed in this paper based on a self-organizing recurrent radial basis function (SR-RBF) neural network, whose structure and parameters are adjusted concurrently in the training process. The proposed SR-RBF neural network is represented in a general nonlinear form for predicting the future dynamic behaviors of nonlinear systems. To improve the modeling accuracy, a spiking-based growing and pruning algorithm and an adaptive learning algorithm are developed to tune the structure and parameters of the SR-RBF neural network, respectively. Meanwhile, for the control problem, an improved gradient method is utilized for the solution of the optimization problem in NMPC. The stability of the resulting control system is proved based on the Lyapunov stability theory. Finally, the proposed SR-RBF neural network-based NMPC (SR-RBF-NMPC) is used to control the dissolved oxygen (DO) concentration in a wastewater treatment process (WWTP). Comparisons with other existing methods demonstrate that the SR-RBF-NMPC can achieve a considerably better model fitting for WWTP and a better control performance for DO concentration.

Fuzzy Counter Propagation Neural Network Control for a Class of Nonlinear Dynamical Systems.

Science.gov (United States)

Sakhre, Vandana; Jain, Sanjeev; Sapkal, Vilas S; Agarwal, Dev P

2015-01-01

Fuzzy Counter Propagation Neural Network (FCPN) controller design is developed, for a class of nonlinear dynamical systems. In this process, the weight connecting between the instar and outstar, that is, input-hidden and hidden-output layer, respectively, is adjusted by using Fuzzy Competitive Learning (FCL). FCL paradigm adopts the principle of learning, which is used to calculate Best Matched Node (BMN) which is proposed. This strategy offers a robust control of nonlinear dynamical systems. FCPN is compared with the existing network like Dynamic Network (DN) and Back Propagation Network (BPN) on the basis of Mean Absolute Error (MAE), Mean Square Error (MSE), Best Fit Rate (BFR), and so forth. It envisages that the proposed FCPN gives better results than DN and BPN. The effectiveness of the proposed FCPN algorithms is demonstrated through simulations of four nonlinear dynamical systems and multiple input and single output (MISO) and a single input and single output (SISO) gas furnace Box-Jenkins time series data.

Fuzzy Counter Propagation Neural Network Control for a Class of Nonlinear Dynamical Systems

Directory of Open Access Journals (Sweden)

Vandana Sakhre

2015-01-01

Full Text Available Fuzzy Counter Propagation Neural Network (FCPN controller design is developed, for a class of nonlinear dynamical systems. In this process, the weight connecting between the instar and outstar, that is, input-hidden and hidden-output layer, respectively, is adjusted by using Fuzzy Competitive Learning (FCL. FCL paradigm adopts the principle of learning, which is used to calculate Best Matched Node (BMN which is proposed. This strategy offers a robust control of nonlinear dynamical systems. FCPN is compared with the existing network like Dynamic Network (DN and Back Propagation Network (BPN on the basis of Mean Absolute Error (MAE, Mean Square Error (MSE, Best Fit Rate (BFR, and so forth. It envisages that the proposed FCPN gives better results than DN and BPN. The effectiveness of the proposed FCPN algorithms is demonstrated through simulations of four nonlinear dynamical systems and multiple input and single output (MISO and a single input and single output (SISO gas furnace Box-Jenkins time series data.

Spatially Nonlinear Interdependence of Alpha-Oscillatory Neural Networks under Chan Meditation

Directory of Open Access Journals (Sweden)

Pei-Chen Lo

2013-01-01

Full Text Available This paper reports the results of our investigation of the effects of Chan meditation on brain electrophysiological behaviors from the viewpoint of spatially nonlinear interdependence among regional neural networks. Particular emphasis is laid on the alpha-dominated EEG (electroencephalograph. Continuous-time wavelet transform was adopted to detect the epochs containing substantial alpha activities. Nonlinear interdependence quantified by similarity index S(X∣Y, the influence of source signal Y on sink signal X, was applied to the nonlinear dynamical model in phase space reconstructed from multichannel EEG. Experimental group involved ten experienced Chan-Meditation practitioners, while control group included ten healthy subjects within the same age range, yet, without any meditation experience. Nonlinear interdependence among various cortical regions was explored for five local neural-network regions, frontal, posterior, right-temporal, left-temporal, and central regions. In the experimental group, the inter-regional interaction was evaluated for the brain dynamics under three different stages, at rest (stage R, pre-meditation background recording, in Chan meditation (stage M, and the unique Chakra-focusing practice (stage C. Experimental group exhibits stronger interactions among various local neural networks at stages M and C compared with those at stage R. The intergroup comparison demonstrates that Chan-meditation brain possesses better cortical inter-regional interactions than the resting brain of control group.

Digital mammographic tumor classification using transfer learning from deep convolutional neural networks.

Science.gov (United States)

Huynh, Benjamin Q; Li, Hui; Giger, Maryellen L

2016-07-01

Convolutional neural networks (CNNs) show potential for computer-aided diagnosis (CADx) by learning features directly from the image data instead of using analytically extracted features. However, CNNs are difficult to train from scratch for medical images due to small sample sizes and variations in tumor presentations. Instead, transfer learning can be used to extract tumor information from medical images via CNNs originally pretrained for nonmedical tasks, alleviating the need for large datasets. Our database includes 219 breast lesions (607 full-field digital mammographic images). We compared support vector machine classifiers based on the CNN-extracted image features and our prior computer-extracted tumor features in the task of distinguishing between benign and malignant breast lesions. Five-fold cross validation (by lesion) was conducted with the area under the receiver operating characteristic (ROC) curve as the performance metric. Results show that classifiers based on CNN-extracted features (with transfer learning) perform comparably to those using analytically extracted features [area under the ROC curve [Formula: see text

Nonlinear signaling on biological networks: The role of stochasticity and spectral clustering

Science.gov (United States)

Hernandez-Hernandez, Gonzalo; Myers, Jesse; Alvarez-Lacalle, Enrique; Shiferaw, Yohannes

2017-03-01

Signal transduction within biological cells is governed by networks of interacting proteins. Communication between these proteins is mediated by signaling molecules which bind to receptors and induce stochastic transitions between different conformational states. Signaling is typically a cooperative process which requires the occurrence of multiple binding events so that reaction rates have a nonlinear dependence on the amount of signaling molecule. It is this nonlinearity that endows biological signaling networks with robust switchlike properties which are critical to their biological function. In this study we investigate how the properties of these signaling systems depend on the network architecture. Our main result is that these nonlinear networks exhibit bistability where the network activity can switch between states that correspond to a low and high activity level. We show that this bistable regime emerges at a critical coupling strength that is determined by the spectral structure of the network. In particular, the set of nodes that correspond to large components of the leading eigenvector of the adjacency matrix determines the onset of bistability. Above this transition the eigenvectors of the adjacency matrix determine a hierarchy of clusters, defined by its spectral properties, which are activated sequentially with increasing network activity. We argue further that the onset of bistability occurs either continuously or discontinuously depending upon whether the leading eigenvector is localized or delocalized. Finally, we show that at low network coupling stochastic transitions to the active branch are also driven by the set of nodes that contribute more strongly to the leading eigenvector. However, at high coupling, transitions are insensitive to network structure since the network can be activated by stochastic transitions of a few nodes. Thus this work identifies important features of biological signaling networks that may underlie their biological

Computing with networks of nonlinear mechanical oscillators.

Directory of Open Access Journals (Sweden)

Jean C Coulombe

Full Text Available As it is getting increasingly difficult to achieve gains in the density and power efficiency of microelectronic computing devices because of lithographic techniques reaching fundamental physical limits, new approaches are required to maximize the benefits of distributed sensors, micro-robots or smart materials. Biologically-inspired devices, such as artificial neural networks, can process information with a high level of parallelism to efficiently solve difficult problems, even when implemented using conventional microelectronic technologies. We describe a mechanical device, which operates in a manner similar to artificial neural networks, to solve efficiently two difficult benchmark problems (computing the parity of a bit stream, and classifying spoken words. The device consists in a network of masses coupled by linear springs and attached to a substrate by non-linear springs, thus forming a network of anharmonic oscillators. As the masses can directly couple to forces applied on the device, this approach combines sensing and computing functions in a single power-efficient device with compact dimensions.

Neural network approximation of nonlinearity in laser nano-metrology system based on TLMI

Energy Technology Data Exchange (ETDEWEB)

Olyaee, Saeed; Hamedi, Samaneh, E-mail: s_olyaee@srttu.edu [Nano-photonics and Optoelectronics Research Laboratory (NORLab), Faculty of Electrical and Computer Engineering, Shahid Rajaee Teacher Training University (SRTTU), Lavizan, 16788, Tehran (Iran, Islamic Republic of)

2011-02-01

In this paper, an approach based on neural network (NN) for nonlinearity modeling in a nano-metrology system using three-longitudinal-mode laser heterodyne interferometer (TLMI) for length and displacement measurements is presented. We model nonlinearity errors that arise from elliptically and non-orthogonally polarized laser beams, rotational error in the alignment of laser head with respect to the polarizing beam splitter, rotational error in the alignment of the mixing polarizer, and unequal transmission coefficients in the polarizing beam splitter. Here we use a neural network algorithm based on the multi-layer perceptron (MLP) network. The simulation results show that multi-layer feed forward perceptron network is successfully applicable to real noisy interferometer signals.

Complex fluid network optimization and control integrative design based on nonlinear dynamic model

International Nuclear Information System (INIS)

Sui, Jinxue; Yang, Li; Hu, Yunan

2016-01-01

In view of distribution according to complex fluid network’s needs, this paper proposed one optimization computation method of the nonlinear programming mathematical model based on genetic algorithm. The simulation result shows that the overall energy consumption of the optimized fluid network has a decrease obviously. The control model of the fluid network is established based on nonlinear dynamics. We design the control law based on feedback linearization, take the optimal value by genetic algorithm as the simulation data, can also solve the branch resistance under the optimal value. These resistances can provide technical support and reference for fluid network design and construction, so can realize complex fluid network optimization and control integration design.

Neural network approximation of nonlinearity in laser nano-metrology system based on TLMI

International Nuclear Information System (INIS)

Olyaee, Saeed; Hamedi, Samaneh

2011-01-01

In this paper, an approach based on neural network (NN) for nonlinearity modeling in a nano-metrology system using three-longitudinal-mode laser heterodyne interferometer (TLMI) for length and displacement measurements is presented. We model nonlinearity errors that arise from elliptically and non-orthogonally polarized laser beams, rotational error in the alignment of laser head with respect to the polarizing beam splitter, rotational error in the alignment of the mixing polarizer, and unequal transmission coefficients in the polarizing beam splitter. Here we use a neural network algorithm based on the multi-layer perceptron (MLP) network. The simulation results show that multi-layer feed forward perceptron network is successfully applicable to real noisy interferometer signals.

Nonlinear System Identification Using Neural Networks Trained with Natural Gradient Descent

Directory of Open Access Journals (Sweden)

Ibnkahla Mohamed

2003-01-01

Full Text Available We use natural gradient (NG learning neural networks (NNs for modeling and identifying nonlinear systems with memory. The nonlinear system is comprised of a discrete-time linear filter followed by a zero-memory nonlinearity . The NN model is composed of a linear adaptive filter followed by a two-layer memoryless nonlinear NN. A Kalman filter-based technique and a search-and-converge method have been employed for the NG algorithm. It is shown that the NG descent learning significantly outperforms the ordinary gradient descent and the Levenberg-Marquardt (LM procedure in terms of convergence speed and mean squared error (MSE performance.

Monitoring tool usage in surgery videos using boosted convolutional and recurrent neural networks.

Science.gov (United States)

Al Hajj, Hassan; Lamard, Mathieu; Conze, Pierre-Henri; Cochener, Béatrice; Quellec, Gwenolé

2018-05-09

This paper investigates the automatic monitoring of tool usage during a surgery, with potential applications in report generation, surgical training and real-time decision support. Two surgeries are considered: cataract surgery, the most common surgical procedure, and cholecystectomy, one of the most common digestive surgeries. Tool usage is monitored in videos recorded either through a microscope (cataract surgery) or an endoscope (cholecystectomy). Following state-of-the-art video analysis solutions, each frame of the video is analyzed by convolutional neural networks (CNNs) whose outputs are fed to recurrent neural networks (RNNs) in order to take temporal relationships between events into account. Novelty lies in the way those CNNs and RNNs are trained. Computational complexity prevents the end-to-end training of "CNN+RNN" systems. Therefore, CNNs are usually trained first, independently from the RNNs. This approach is clearly suboptimal for surgical tool analysis: many tools are very similar to one another, but they can generally be differentiated based on past events. CNNs should be trained to extract the most useful visual features in combination with the temporal context. A novel boosting strategy is proposed to achieve this goal: the CNN and RNN parts of the system are simultaneously enriched by progressively adding weak classifiers (either CNNs or RNNs) trained to improve the overall classification accuracy. Experiments were performed in a dataset of 50 cataract surgery videos, where the usage of 21 surgical tools was manually annotated, and a dataset of 80 cholecystectomy videos, where the usage of 7 tools was manually annotated. Very good classification performance are achieved in both datasets: tool usage could be labeled with an average area under the ROC curve of A z =0.9961 and A z =0.9939, respectively, in offline mode (using past, present and future information), and A z =0.9957 and A z =0.9936, respectively, in online mode (using past and present

Mean Square Synchronization of Stochastic Nonlinear Delayed Coupled Complex Networks

Directory of Open Access Journals (Sweden)

Chengrong Xie

2013-01-01

Full Text Available We investigate the problem of adaptive mean square synchronization for nonlinear delayed coupled complex networks with stochastic perturbation. Based on the LaSalle invariance principle and the properties of the Weiner process, the controller and adaptive laws are designed to ensure achieving stochastic synchronization and topology identification of complex networks. Sufficient conditions are given to ensure the complex networks to be mean square synchronization. Furthermore, numerical simulations are also given to demonstrate the effectiveness of the proposed scheme.

Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection.

Science.gov (United States)

Kim, Jihun; Kim, Jonghong; Jang, Gil-Jin; Lee, Minho

2017-03-01

Deep learning has received significant attention recently as a promising solution to many problems in the area of artificial intelligence. Among several deep learning architectures, convolutional neural networks (CNNs) demonstrate superior performance when compared to other machine learning methods in the applications of object detection and recognition. We use a CNN for image enhancement and the detection of driving lanes on motorways. In general, the process of lane detection consists of edge extraction and line detection. A CNN can be used to enhance the input images before lane detection by excluding noise and obstacles that are irrelevant to the edge detection result. However, training conventional CNNs requires considerable computation and a big dataset. Therefore, we suggest a new learning algorithm for CNNs using an extreme learning machine (ELM). The ELM is a fast learning method used to calculate network weights between output and hidden layers in a single iteration and thus, can dramatically reduce learning time while producing accurate results with minimal training data. A conventional ELM can be applied to networks with a single hidden layer; as such, we propose a stacked ELM architecture in the CNN framework. Further, we modify the backpropagation algorithm to find the targets of hidden layers and effectively learn network weights while maintaining performance. Experimental results confirm that the proposed method is effective in reducing learning time and improving performance. Copyright © 2016 Elsevier Ltd. All rights reserved.

Robust output synchronization of heterogeneous nonlinear agents in uncertain networks.

Science.gov (United States)

Yang, Xi; Wan, Fuhua; Tu, Mengchuan; Shen, Guojiang

2017-11-01

This paper investigates the global robust output synchronization problem for a class of nonlinear multi-agent systems. In the considered setup, the controlled agents are heterogeneous and with both dynamic and parametric uncertainties, the controllers are incapable of exchanging their internal states with the neighbors, and the communication network among agents is defined by an uncertain simple digraph. The problem is pursued via nonlinear output regulation theory and internal model based design. For each agent, the input-driven filter and the internal model compose the controller, and the decentralized dynamic output feedback control law is derived by using backstepping method and the modified dynamic high-gain technique. The theoretical result is applied to output synchronization problem for uncertain network of Lorenz-type agents. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.

An integer optimization algorithm for robust identification of non-linear gene regulatory networks

Directory of Open Access Journals (Sweden)

Chemmangattuvalappil Nishanth

2012-09-01

Full Text Available Abstract Background Reverse engineering gene networks and identifying regulatory interactions are integral to understanding cellular decision making processes. Advancement in high throughput experimental techniques has initiated innovative data driven analysis of gene regulatory networks. However, inherent noise associated with biological systems requires numerous experimental replicates for reliable conclusions. Furthermore, evidence of robust algorithms directly exploiting basic biological traits are few. Such algorithms are expected to be efficient in their performance and robust in their prediction. Results We have developed a network identification algorithm to accurately infer both the topology and strength of regulatory interactions from time series gene expression data in the presence of significant experimental noise and non-linear behavior. In this novel formulism, we have addressed data variability in biological systems by integrating network identification with the bootstrap resampling technique, hence predicting robust interactions from limited experimental replicates subjected to noise. Furthermore, we have incorporated non-linearity in gene dynamics using the S-system formulation. The basic network identification formulation exploits the trait of sparsity of biological interactions. Towards that, the identification algorithm is formulated as an integer-programming problem by introducing binary variables for each network component. The objective function is targeted to minimize the network connections subjected to the constraint of maximal agreement between the experimental and predicted gene dynamics. The developed algorithm is validated using both in silico and experimental data-sets. These studies show that the algorithm can accurately predict the topology and connection strength of the in silico networks, as quantified by high precision and recall, and small discrepancy between the actual and predicted kinetic parameters

Inferring nonlinear gene regulatory networks from gene expression data based on distance correlation.

Directory of Open Access Journals (Sweden)

Xiaobo Guo

Full Text Available Nonlinear dependence is general in regulation mechanism of gene regulatory networks (GRNs. It is vital to properly measure or test nonlinear dependence from real data for reconstructing GRNs and understanding the complex regulatory mechanisms within the cellular system. A recently developed measurement called the distance correlation (DC has been shown powerful and computationally effective in nonlinear dependence for many situations. In this work, we incorporate the DC into inferring GRNs from the gene expression data without any underling distribution assumptions. We propose three DC-based GRNs inference algorithms: CLR-DC, MRNET-DC and REL-DC, and then compare them with the mutual information (MI-based algorithms by analyzing two simulated data: benchmark GRNs from the DREAM challenge and GRNs generated by SynTReN network generator, and an experimentally determined SOS DNA repair network in Escherichia coli. According to both the receiver operator characteristic (ROC curve and the precision-recall (PR curve, our proposed algorithms significantly outperform the MI-based algorithms in GRNs inference.

Computationally Efficient Nonlinear Bell Inequalities for Quantum Networks

Science.gov (United States)

Luo, Ming-Xing

2018-04-01

The correlations in quantum networks have attracted strong interest with new types of violations of the locality. The standard Bell inequalities cannot characterize the multipartite correlations that are generated by multiple sources. The main problem is that no computationally efficient method is available for constructing useful Bell inequalities for general quantum networks. In this work, we show a significant improvement by presenting new, explicit Bell-type inequalities for general networks including cyclic networks. These nonlinear inequalities are related to the matching problem of an equivalent unweighted bipartite graph that allows constructing a polynomial-time algorithm. For the quantum resources consisting of bipartite entangled pure states and generalized Greenberger-Horne-Zeilinger (GHZ) states, we prove the generic nonmultilocality of quantum networks with multiple independent observers using new Bell inequalities. The violations are maximal with respect to the presented Tsirelson's bound for Einstein-Podolsky-Rosen states and GHZ states. Moreover, these violations hold for Werner states or some general noisy states. Our results suggest that the presented Bell inequalities can be used to characterize experimental quantum networks.

Neural feedback linearization adaptive control for affine nonlinear systems based on neural network estimator

Directory of Open Access Journals (Sweden)

Bahita Mohamed

2011-01-01

Full Text Available In this work, we introduce an adaptive neural network controller for a class of nonlinear systems. The approach uses two Radial Basis Functions, RBF networks. The first RBF network is used to approximate the ideal control law which cannot be implemented since the dynamics of the system are unknown. The second RBF network is used for on-line estimating the control gain which is a nonlinear and unknown function of the states. The updating laws for the combined estimator and controller are derived through Lyapunov analysis. Asymptotic stability is established with the tracking errors converging to a neighborhood of the origin. Finally, the proposed method is applied to control and stabilize the inverted pendulum system.

Non-linear contributions to interactions in climate networks: sources, relevance, nonstationarity

Czech Academy of Sciences Publication Activity Database

Hlinka, Jaroslav; Hartman, David; Vejmelka, Martin; Paluš, Milan

2012-01-01

Roč. 14, - (2012), s. 14274 ISSN 1607-7962. [European Geosciences Union General Assembly 2012. 22.04.2012-27.04.2012, Vienna] R&D Projects: GA ČR GCP103/11/J068 Institutional support: RVO:67985807 Keywords : correlation * mutual information * test of nonlinearity * surrogate data * complex networks * climate network Subject RIV: BB - Applied Statistics, Operational Research

Location Sensitive Deep Convolutional Neural Networks for Segmentation of White Matter Hyperintensities.

Science.gov (United States)

Ghafoorian, Mohsen; Karssemeijer, Nico; Heskes, Tom; van Uden, Inge W M; Sanchez, Clara I; Litjens, Geert; de Leeuw, Frank-Erik; van Ginneken, Bram; Marchiori, Elena; Platel, Bram

2017-07-11

The anatomical location of imaging features is of crucial importance for accurate diagnosis in many medical tasks. Convolutional neural networks (CNN) have had huge successes in computer vision, but they lack the natural ability to incorporate the anatomical location in their decision making process, hindering success in some medical image analysis tasks. In this paper, to integrate the anatomical location information into the network, we propose several deep CNN architectures that consider multi-scale patches or take explicit location features while training. We apply and compare the proposed architectures for segmentation of white matter hyperintensities in brain MR images on a large dataset. As a result, we observe that the CNNs that incorporate location information substantially outperform a conventional segmentation method with handcrafted features as well as CNNs that do not integrate location information. On a test set of 50 scans, the best configuration of our networks obtained a Dice score of 0.792, compared to 0.805 for an independent human observer. Performance levels of the machine and the independent human observer were not statistically significantly different (p-value = 0.06).

Nonlinear neural network for hemodynamic model state and input estimation using fMRI data

KAUST Repository

Karam, Ayman M.

2014-11-01

Originally inspired by biological neural networks, artificial neural networks (ANNs) are powerful mathematical tools that can solve complex nonlinear problems such as filtering, classification, prediction and more. This paper demonstrates the first successful implementation of ANN, specifically nonlinear autoregressive with exogenous input (NARX) networks, to estimate the hemodynamic states and neural activity from simulated and measured real blood oxygenation level dependent (BOLD) signals. Blocked and event-related BOLD data are used to test the algorithm on real experiments. The proposed method is accurate and robust even in the presence of signal noise and it does not depend on sampling interval. Moreover, the structure of the NARX networks is optimized to yield the best estimate with minimal network architecture. The results of the estimated neural activity are also discussed in terms of their potential use.

Unified functional network and nonlinear time series analysis for complex systems science: The pyunicorn package

Science.gov (United States)

Donges, Jonathan; Heitzig, Jobst; Beronov, Boyan; Wiedermann, Marc; Runge, Jakob; Feng, Qing Yi; Tupikina, Liubov; Stolbova, Veronika; Donner, Reik; Marwan, Norbert; Dijkstra, Henk; Kurths, Jürgen

2016-04-01

We introduce the pyunicorn (Pythonic unified complex network and recurrence analysis toolbox) open source software package for applying and combining modern methods of data analysis and modeling from complex network theory and nonlinear time series analysis. pyunicorn is a fully object-oriented and easily parallelizable package written in the language Python. It allows for the construction of functional networks such as climate networks in climatology or functional brain networks in neuroscience representing the structure of statistical interrelationships in large data sets of time series and, subsequently, investigating this structure using advanced methods of complex network theory such as measures and models for spatial networks, networks of interacting networks, node-weighted statistics, or network surrogates. Additionally, pyunicorn provides insights into the nonlinear dynamics of complex systems as recorded in uni- and multivariate time series from a non-traditional perspective by means of recurrence quantification analysis, recurrence networks, visibility graphs, and construction of surrogate time series. The range of possible applications of the library is outlined, drawing on several examples mainly from the field of climatology. pyunicorn is available online at https://github.com/pik-copan/pyunicorn. Reference: J.F. Donges, J. Heitzig, B. Beronov, M. Wiedermann, J. Runge, Q.-Y. Feng, L. Tupikina, V. Stolbova, R.V. Donner, N. Marwan, H.A. Dijkstra, and J. Kurths, Unified functional network and nonlinear time series analysis for complex systems science: The pyunicorn package, Chaos 25, 113101 (2015), DOI: 10.1063/1.4934554, Preprint: arxiv.org:1507.01571 [physics.data-an].

Path selection and bandwidth allocation in MPLS networks: a nonlinear programming approach

Science.gov (United States)

Burns, J. E.; Ott, Teunis J.; de Kock, Johan M.; Krzesinski, Anthony E.

2001-07-01

Multi-protocol Label Switching extends the IPv4 destination-based routing protocols to provide new and scalable routing capabilities in connectionless networks using relatively simple packet forwarding mechanisms. MPLS networks carry traffic on virtual connections called label switched paths. This paper considers path selection and bandwidth allocation in MPLS networks in order to optimize the network quality of service. The optimization is based upon the minimization of a non-linear objective function which under light load simplifies to OSPF routing with link metrics equal to the link propagation delays. The behavior under heavy load depends on the choice of certain parameters: It can essentially be made to minimize maximal expected utilization, or to maximize minimal expected weighted slacks (both over all links). Under certain circumstances it can be made to minimize the probability that a link has an instantaneous offered load larger than its transmission capacity. We present a model of an MPLS network and an algorithm to find and capacitate optimal LSPs. The algorithm is an improvement of the well-known flow deviation non-linear programming method. The algorithm is applied to compute optimal LSPs for several test networks carrying a single traffic class.

A light and faster regional convolutional neural network for object detection in optical remote sensing images

Science.gov (United States)

Ding, Peng; Zhang, Ye; Deng, Wei-Jian; Jia, Ping; Kuijper, Arjan

2018-07-01

Detection of objects from satellite optical remote sensing images is very important for many commercial and governmental applications. With the development of deep convolutional neural networks (deep CNNs), the field of object detection has seen tremendous advances. Currently, objects in satellite remote sensing images can be detected using deep CNNs. In general, optical remote sensing images contain many dense and small objects, and the use of the original Faster Regional CNN framework does not yield a suitably high precision. Therefore, after careful analysis we adopt dense convoluted networks, a multi-scale representation and various combinations of improvement schemes to enhance the structure of the base VGG16-Net for improving the precision. We propose an approach to reduce the test-time (detection time) and memory requirements. To validate the effectiveness of our approach, we perform experiments using satellite remote sensing image datasets of aircraft and automobiles. The results show that the improved network structure can detect objects in satellite optical remote sensing images more accurately and efficiently.

Deep Convolutional Neural Networks for Multi-Modality Isointense Infant Brain Image Segmentation

Science.gov (United States)

Zhang, Wenlu; Li, Rongjian; Deng, Houtao; Wang, Li; Lin, Weili; Ji, Shuiwang; Shen, Dinggang

2015-01-01

The segmentation of infant brain tissue images into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) plays an important role in studying early brain development in health and disease. In the isointense stage (approximately 6–8 months of age), WM and GM exhibit similar levels of intensity in both T1 and T2 MR images, making the tissue segmentation very challenging. Only a small number of existing methods have been designed for tissue segmentation in this isointense stage; however, they only used a single T1 or T2 images, or the combination of T1 and T2 images. In this paper, we propose to use deep convolutional neural networks (CNNs) for segmenting isointense stage brain tissues using multi-modality MR images. CNNs are a type of deep models in which trainable filters and local neighborhood pooling operations are applied alternatingly on the raw input images, resulting in a hierarchy of increasingly complex features. Specifically, we used multimodality information from T1, T2, and fractional anisotropy (FA) images as inputs and then generated the segmentation maps as outputs. The multiple intermediate layers applied convolution, pooling, normalization, and other operations to capture the highly nonlinear mappings between inputs and outputs. We compared the performance of our approach with that of the commonly used segmentation methods on a set of manually segmented isointense stage brain images. Results showed that our proposed model significantly outperformed prior methods on infant brain tissue segmentation. In addition, our results indicated that integration of multi-modality images led to significant performance improvement. PMID:25562829

Synthetic bootstrapping of convolutional neural networks for semantic plant part segmentation

NARCIS (Netherlands)

Barth, R.; IJsselmuiden, J.; Hemming, J.; Henten, Van E.J.

2017-01-01

A current bottleneck of state-of-the-art machine learning methods for image segmentation in agriculture, e.g. convolutional neural networks (CNNs), is the requirement of large manually annotated datasets on a per-pixel level. In this paper, we investigated how related synthetic images can be used to

Real Time Eye Detector with Cascaded Convolutional Neural Networks

Directory of Open Access Journals (Sweden)

Bin Li

2018-01-01

Full Text Available An accurate and efficient eye detector is essential for many computer vision applications. In this paper, we present an efficient method to evaluate the eye location from facial images. First, a group of candidate regions with regional extreme points is quickly proposed; then, a set of convolution neural networks (CNNs is adopted to determine the most likely eye region and classify the region as left or right eye; finally, the center of the eye is located with other CNNs. In the experiments using GI4E, BioID, and our datasets, our method attained a detection accuracy which is comparable to existing state-of-the-art methods; meanwhile, our method was faster and adaptable to variations of the images, including external light changes, facial occlusion, and changes in image modality.

Adaptive Control of Nonlinear Discrete-Time Systems by Using OS-ELM Neural Networks

Directory of Open Access Journals (Sweden)

Xiao-Li Li

2014-01-01

Full Text Available As a kind of novel feedforward neural network with single hidden layer, ELM (extreme learning machine neural networks are studied for the identification and control of nonlinear dynamic systems. The property of simple structure and fast convergence of ELM can be shown clearly. In this paper, we are interested in adaptive control of nonlinear dynamic plants by using OS-ELM (online sequential extreme learning machine neural networks. Based on data scope division, the problem that training process of ELM neural network is sensitive to the initial training data is also solved. According to the output range of the controlled plant, the data corresponding to this range will be used to initialize ELM. Furthermore, due to the drawback of conventional adaptive control, when the OS-ELM neural network is used for adaptive control of the system with jumping parameters, the topological structure of the neural network can be adjusted dynamically by using multiple model switching strategy, and an MMAC (multiple model adaptive control will be used to improve the control performance. Simulation results are included to complement the theoretical results.

Neural networks for feedback feedforward nonlinear control systems.

Science.gov (United States)

Parisini, T; Zoppoli, R

1994-01-01

This paper deals with the problem of designing feedback feedforward control strategies to drive the state of a dynamic system (in general, nonlinear) so as to track any desired trajectory joining the points of given compact sets, while minimizing a certain cost function (in general, nonquadratic). Due to the generality of the problem, conventional methods are difficult to apply. Thus, an approximate solution is sought by constraining control strategies to take on the structure of multilayer feedforward neural networks. After discussing the approximation properties of neural control strategies, a particular neural architecture is presented, which is based on what has been called the "linear-structure preserving principle". The original functional problem is then reduced to a nonlinear programming one, and backpropagation is applied to derive the optimal values of the synaptic weights. Recursive equations to compute the gradient components are presented, which generalize the classical adjoint system equations of N-stage optimal control theory. Simulation results related to nonlinear nonquadratic problems show the effectiveness of the proposed method.

Deep Recurrent Neural Networks for Human Activity Recognition

Directory of Open Access Journals (Sweden)

Abdulmajid Murad

2017-11-01

Full Text Available Adopting deep learning methods for human activity recognition has been effective in extracting discriminative features from raw input sequences acquired from body-worn sensors. Although human movements are encoded in a sequence of successive samples in time, typical machine learning methods perform recognition tasks without exploiting the temporal correlations between input data samples. Convolutional neural networks (CNNs address this issue by using convolutions across a one-dimensional temporal sequence to capture dependencies among input data. However, the size of convolutional kernels restricts the captured range of dependencies between data samples. As a result, typical models are unadaptable to a wide range of activity-recognition configurations and require fixed-length input windows. In this paper, we propose the use of deep recurrent neural networks (DRNNs for building recognition models that are capable of capturing long-range dependencies in variable-length input sequences. We present unidirectional, bidirectional, and cascaded architectures based on long short-term memory (LSTM DRNNs and evaluate their effectiveness on miscellaneous benchmark datasets. Experimental results show that our proposed models outperform methods employing conventional machine learning, such as support vector machine (SVM and k-nearest neighbors (KNN. Additionally, the proposed models yield better performance than other deep learning techniques, such as deep believe networks (DBNs and CNNs.

Deep Recurrent Neural Networks for Human Activity Recognition.

Science.gov (United States)

Murad, Abdulmajid; Pyun, Jae-Young

2017-11-06

Adopting deep learning methods for human activity recognition has been effective in extracting discriminative features from raw input sequences acquired from body-worn sensors. Although human movements are encoded in a sequence of successive samples in time, typical machine learning methods perform recognition tasks without exploiting the temporal correlations between input data samples. Convolutional neural networks (CNNs) address this issue by using convolutions across a one-dimensional temporal sequence to capture dependencies among input data. However, the size of convolutional kernels restricts the captured range of dependencies between data samples. As a result, typical models are unadaptable to a wide range of activity-recognition configurations and require fixed-length input windows. In this paper, we propose the use of deep recurrent neural networks (DRNNs) for building recognition models that are capable of capturing long-range dependencies in variable-length input sequences. We present unidirectional, bidirectional, and cascaded architectures based on long short-term memory (LSTM) DRNNs and evaluate their effectiveness on miscellaneous benchmark datasets. Experimental results show that our proposed models outperform methods employing conventional machine learning, such as support vector machine (SVM) and k-nearest neighbors (KNN). Additionally, the proposed models yield better performance than other deep learning techniques, such as deep believe networks (DBNs) and CNNs.

Mutual Connectivity Analysis (MCA) Using Generalized Radial Basis Function Neural Networks for Nonlinear Functional Connectivity Network Recovery in Resting-State Functional MRI.

Science.gov (United States)

DSouza, Adora M; Abidin, Anas Zainul; Nagarajan, Mahesh B; Wismüller, Axel

2016-03-29

We investigate the applicability of a computational framework, called mutual connectivity analysis (MCA), for directed functional connectivity analysis in both synthetic and resting-state functional MRI data. This framework comprises of first evaluating non-linear cross-predictability between every pair of time series prior to recovering the underlying network structure using community detection algorithms. We obtain the non-linear cross-prediction score between time series using Generalized Radial Basis Functions (GRBF) neural networks. These cross-prediction scores characterize the underlying functionally connected networks within the resting brain, which can be extracted using non-metric clustering approaches, such as the Louvain method. We first test our approach on synthetic models with known directional influence and network structure. Our method is able to capture the directional relationships between time series (with an area under the ROC curve = 0.92 ± 0.037) as well as the underlying network structure (Rand index = 0.87 ± 0.063) with high accuracy. Furthermore, we test this method for network recovery on resting-state fMRI data, where results are compared to the motor cortex network recovered from a motor stimulation sequence, resulting in a strong agreement between the two (Dice coefficient = 0.45). We conclude that our MCA approach is effective in analyzing non-linear directed functional connectivity and in revealing underlying functional network structure in complex systems.

Correlation dimension based nonlinear analysis of network traffics with different application protocols

International Nuclear Information System (INIS)

Wang Jun-Song; Yuan Jing; Li Qiang; Yuan Rui-Xi

2011-01-01

This paper uses a correlation dimension based nonlinear analysis approach to analyse the dynamics of network traffics with three different application protocols—HTTP, FTP and SMTP. First, the phase space is reconstructed and the embedding parameters are obtained by the mutual information method. Secondly, the correlation dimensions of three different traffics are calculated and the results of analysis have demonstrated that the dynamics of the three different application protocol traffics is different from each other in nature, i.e. HTTP and FTP traffics are chaotic, furthermore, the former is more complex than the later; on the other hand, SMTP traffic is stochastic. It is shown that correlation dimension approach is an efficient method to understand and to characterize the nonlinear dynamics of HTTP, FTP and SMTP protocol network traffics. This analysis provided insight into and a more accurate understanding of nonlinear dynamics of internet traffics which have a complex mixture of chaotic and stochastic components. (general)

Wireless and real-time structural damage detection: A novel decentralized method for wireless sensor networks

Science.gov (United States)

Avci, Onur; Abdeljaber, Osama; Kiranyaz, Serkan; Hussein, Mohammed; Inman, Daniel J.

2018-06-01

Being an alternative to conventional wired sensors, wireless sensor networks (WSNs) are extensively used in Structural Health Monitoring (SHM) applications. Most of the Structural Damage Detection (SDD) approaches available in the SHM literature are centralized as they require transferring data from all sensors within the network to a single processing unit to evaluate the structural condition. These methods are found predominantly feasible for wired SHM systems; however, transmission and synchronization of huge data sets in WSNs has been found to be arduous. As such, the application of centralized methods with WSNs has been a challenge for engineers. In this paper, the authors are presenting a novel application of 1D Convolutional Neural Networks (1D CNNs) on WSNs for SDD purposes. The SDD is successfully performed completely wireless and real-time under ambient conditions. As a result of this, a decentralized damage detection method suitable for wireless SHM systems is proposed. The proposed method is based on 1D CNNs and it involves training an individual 1D CNN for each wireless sensor in the network in a format where each CNN is assigned to process the locally-available data only, eliminating the need for data transmission and synchronization. The proposed damage detection method operates directly on the raw ambient vibration condition signals without any filtering or preprocessing. Moreover, the proposed approach requires minimal computational time and power since 1D CNNs merge both feature extraction and classification tasks into a single learning block. This ability is prevailingly cost-effective and evidently practical in WSNs considering the hardware systems have been occasionally reported to suffer from limited power supply in these networks. To display the capability and verify the success of the proposed method, large-scale experiments conducted on a laboratory structure equipped with a state-of-the-art WSN are reported.

New recursive-least-squares algorithms for nonlinear active control of sound and vibration using neural networks.

Science.gov (United States)

Bouchard, M

2001-01-01

In recent years, a few articles describing the use of neural networks for nonlinear active control of sound and vibration were published. Using a control structure with two multilayer feedforward neural networks (one as a nonlinear controller and one as a nonlinear plant model), steepest descent algorithms based on two distinct gradient approaches were introduced for the training of the controller network. The two gradient approaches were sometimes called the filtered-x approach and the adjoint approach. Some recursive-least-squares algorithms were also introduced, using the adjoint approach. In this paper, an heuristic procedure is introduced for the development of recursive-least-squares algorithms based on the filtered-x and the adjoint gradient approaches. This leads to the development of new recursive-least-squares algorithms for the training of the controller neural network in the two networks structure. These new algorithms produce a better convergence performance than previously published algorithms. Differences in the performance of algorithms using the filtered-x and the adjoint gradient approaches are discussed in the paper. The computational load of the algorithms discussed in the paper is evaluated for multichannel systems of nonlinear active control. Simulation results are presented to compare the convergence performance of the algorithms, showing the convergence gain provided by the new algorithms.

On the synchronization of neural networks containing time-varying delays and sector nonlinearity

International Nuclear Information System (INIS)

Yan, J.-J.; Lin, J.-S.; Hung, M.-L.; Liao, T.-L.

2007-01-01

We present a systematic design procedure for synchronization of neural networks subject to time-varying delays and sector nonlinearity in the control input. Based on the drive-response concept and the Lyapunov stability theorem, a memoryless decentralized control law is proposed which guarantees exponential synchronization even when input nonlinearity is present. The supplementary requirement that the time-derivative of time-varying delays must be smaller than one is released for the proposed control scheme. A four-dimensional Hopfield neural network with time-varying delays is presented as the illustrative example to demonstrate the effectiveness of the proposed synchronization scheme

Structure-based control of complex networks with nonlinear dynamics.

Science.gov (United States)

Zañudo, Jorge Gomez Tejeda; Yang, Gang; Albert, Réka

2017-07-11

What can we learn about controlling a system solely from its underlying network structure? Here we adapt a recently developed framework for control of networks governed by a broad class of nonlinear dynamics that includes the major dynamic models of biological, technological, and social processes. This feedback-based framework provides realizable node overrides that steer a system toward any of its natural long-term dynamic behaviors, regardless of the specific functional forms and system parameters. We use this framework on several real networks, identify the topological characteristics that underlie the predicted node overrides, and compare its predictions to those of structural controllability in control theory. Finally, we demonstrate this framework's applicability in dynamic models of gene regulatory networks and identify nodes whose override is necessary for control in the general case but not in specific model instances.

Nonlinear Robust Observer-Based Fault Detection for Networked Suspension Control System of Maglev Train

Directory of Open Access Journals (Sweden)

Yun Li

2013-01-01

Full Text Available A fault detection approach based on nonlinear robust observer is designed for the networked suspension control system of Maglev train with random induced time delay. First, considering random bounded time-delay and external disturbance, the nonlinear model of the networked suspension control system is established. Then, a nonlinear robust observer is designed using the input of the suspension gap. And the estimate error is proved to be bounded with arbitrary precision by adopting an appropriate parameter. When sensor faults happen, the residual between the real states and the observer outputs indicates which kind of sensor failures occurs. Finally, simulation results using the actual parameters of CMS-04 maglev train indicate that the proposed method is effective for maglev train.

Bearing Fault Diagnosis under Variable Speed Using Convolutional Neural Networks and the Stochastic Diagonal Levenberg-Marquardt Algorithm

Directory of Open Access Journals (Sweden)

Viet Tra

2017-12-01

Full Text Available This paper presents a novel method for diagnosing incipient bearing defects under variable operating speeds using convolutional neural networks (CNNs trained via the stochastic diagonal Levenberg-Marquardt (S-DLM algorithm. The CNNs utilize the spectral energy maps (SEMs of the acoustic emission (AE signals as inputs and automatically learn the optimal features, which yield the best discriminative models for diagnosing incipient bearing defects under variable operating speeds. The SEMs are two-dimensional maps that show the distribution of energy across different bands of the AE spectrum. It is hypothesized that the variation of a bearing’s speed would not alter the overall shape of the AE spectrum rather, it may only scale and translate it. Thus, at different speeds, the same defect would yield SEMs that are scaled and shifted versions of each other. This hypothesis is confirmed by the experimental results, where CNNs trained using the S-DLM algorithm yield significantly better diagnostic performance under variable operating speeds compared to existing methods. In this work, the performance of different training algorithms is also evaluated to select the best training algorithm for the CNNs. The proposed method is used to diagnose both single and compound defects at six different operating speeds.

Differential Neural Networks for Identification and Filtering in Nonlinear Dynamic Games

Directory of Open Access Journals (Sweden)

Emmanuel García

2014-01-01

Full Text Available This paper deals with the problem of identifying and filtering a class of continuous-time nonlinear dynamic games (nonlinear differential games subject to additive and undesired deterministic perturbations. Moreover, the mathematical model of this class is completely unknown with the exception of the control actions of each player, and even though the deterministic noises are known, their power (or their effect is not. Therefore, two differential neural networks are designed in order to obtain a feedback (perfect state information pattern for the mentioned class of games. In this way, the stability conditions for two state identification errors and for a filtering error are established, the upper bounds of these errors are obtained, and two new learning laws for each neural network are suggested. Finally, an illustrating example shows the applicability of this approach.

Deep learning quick reference useful hacks for training and optimizing deep neural networks with TensorFlow and Keras

CERN Document Server

Bernico, Michael

2018-01-01

This book is a practical guide to applying deep neural networks including MLPs, CNNs, LSTMs, and more in Keras and TensorFlow. Packed with useful hacks to solve real-world challenges along with the supported math and theory around each topic, this book will be a quick reference for training and optimize your deep neural networks.

Robust outer synchronization between two nonlinear complex networks with parametric disturbances and mixed time-varying delays

Science.gov (United States)

Zhang, Chuan; Wang, Xingyuan; Luo, Chao; Li, Junqiu; Wang, Chunpeng

2018-03-01

In this paper, we focus on the robust outer synchronization problem between two nonlinear complex networks with parametric disturbances and mixed time-varying delays. Firstly, a general complex network model is proposed. Besides the nonlinear couplings, the network model in this paper can possess parametric disturbances, internal time-varying delay, discrete time-varying delay and distributed time-varying delay. Then, according to the robust control strategy, linear matrix inequality and Lyapunov stability theory, several outer synchronization protocols are strictly derived. Simple linear matrix controllers are designed to driver the response network synchronize to the drive network. Additionally, our results can be applied on the complex networks without parametric disturbances. Finally, by utilizing the delayed Lorenz chaotic system as the dynamics of all nodes, simulation examples are given to demonstrate the effectiveness of our theoretical results.

Sparsity Invariant CNNs

OpenAIRE

Uhrig, Jonas; Schneider, Nick; Schneider, Lukas; Franke, Uwe; Brox, Thomas; Geiger, Andreas

2017-01-01

In this paper, we consider convolutional neural networks operating on sparse inputs with an application to depth upsampling from sparse laser scan data. First, we show that traditional convolutional networks perform poorly when applied to sparse data even when the location of missing data is provided to the network. To overcome this problem, we propose a simple yet effective sparse convolution layer which explicitly considers the location of missing data during the convolution operation. We d...

Visualizing histopathologic deep learning classification and anomaly detection using nonlinear feature space dimensionality reduction.

Science.gov (United States)

Faust, Kevin; Xie, Quin; Han, Dominick; Goyle, Kartikay; Volynskaya, Zoya; Djuric, Ugljesa; Diamandis, Phedias

2018-05-16

There is growing interest in utilizing artificial intelligence, and particularly deep learning, for computer vision in histopathology. While accumulating studies highlight expert-level performance of convolutional neural networks (CNNs) on focused classification tasks, most studies rely on probability distribution scores with empirically defined cutoff values based on post-hoc analysis. More generalizable tools that allow humans to visualize histology-based deep learning inferences and decision making are scarce. Here, we leverage t-distributed Stochastic Neighbor Embedding (t-SNE) to reduce dimensionality and depict how CNNs organize histomorphologic information. Unique to our workflow, we develop a quantitative and transparent approach to visualizing classification decisions prior to softmax compression. By discretizing the relationships between classes on the t-SNE plot, we show we can super-impose randomly sampled regions of test images and use their distribution to render statistically-driven classifications. Therefore, in addition to providing intuitive outputs for human review, this visual approach can carry out automated and objective multi-class classifications similar to more traditional and less-transparent categorical probability distribution scores. Importantly, this novel classification approach is driven by a priori statistically defined cutoffs. It therefore serves as a generalizable classification and anomaly detection tool less reliant on post-hoc tuning. Routine incorporation of this convenient approach for quantitative visualization and error reduction in histopathology aims to accelerate early adoption of CNNs into generalized real-world applications where unanticipated and previously untrained classes are often encountered.

A deep belief network with PLSR for nonlinear system modeling.

Science.gov (United States)

Qiao, Junfei; Wang, Gongming; Li, Wenjing; Li, Xiaoli

2017-10-31

Nonlinear system modeling plays an important role in practical engineering, and deep learning-based deep belief network (DBN) is now popular in nonlinear system modeling and identification because of the strong learning ability. However, the existing weights optimization for DBN is based on gradient, which always leads to a local optimum and a poor training result. In this paper, a DBN with partial least square regression (PLSR-DBN) is proposed for nonlinear system modeling, which focuses on the problem of weights optimization for DBN using PLSR. Firstly, unsupervised contrastive divergence (CD) algorithm is used in weights initialization. Secondly, initial weights derived from CD algorithm are optimized through layer-by-layer PLSR modeling from top layer to bottom layer. Instead of gradient method, PLSR-DBN can determine the optimal weights using several PLSR models, so that a better performance of PLSR-DBN is achieved. Then, the analysis of convergence is theoretically given to guarantee the effectiveness of the proposed PLSR-DBN model. Finally, the proposed PLSR-DBN is tested on two benchmark nonlinear systems and an actual wastewater treatment system as well as a handwritten digit recognition (nonlinear mapping and modeling) with high-dimension input data. The experiment results show that the proposed PLSR-DBN has better performances of time and accuracy on nonlinear system modeling than that of other methods. Copyright © 2017 Elsevier Ltd. All rights reserved.

A novel joint-processing adaptive nonlinear equalizer using a modular recurrent neural network for chaotic communication systems.

Science.gov (United States)

Zhao, Haiquan; Zeng, Xiangping; Zhang, Jiashu; Liu, Yangguang; Wang, Xiaomin; Li, Tianrui

2011-01-01

To eliminate nonlinear channel distortion in chaotic communication systems, a novel joint-processing adaptive nonlinear equalizer based on a pipelined recurrent neural network (JPRNN) is proposed, using a modified real-time recurrent learning (RTRL) algorithm. Furthermore, an adaptive amplitude RTRL algorithm is adopted to overcome the deteriorating effect introduced by the nesting process. Computer simulations illustrate that the proposed equalizer outperforms the pipelined recurrent neural network (PRNN) and recurrent neural network (RNN) equalizers. Copyright © 2010 Elsevier Ltd. All rights reserved.

Robust Vehicle Detection in Aerial Images Based on Cascaded Convolutional Neural Networks.

Science.gov (United States)

Zhong, Jiandan; Lei, Tao; Yao, Guangle

2017-11-24

Vehicle detection in aerial images is an important and challenging task. Traditionally, many target detection models based on sliding-window fashion were developed and achieved acceptable performance, but these models are time-consuming in the detection phase. Recently, with the great success of convolutional neural networks (CNNs) in computer vision, many state-of-the-art detectors have been designed based on deep CNNs. However, these CNN-based detectors are inefficient when applied in aerial image data due to the fact that the existing CNN-based models struggle with small-size object detection and precise localization. To improve the detection accuracy without decreasing speed, we propose a CNN-based detection model combining two independent convolutional neural networks, where the first network is applied to generate a set of vehicle-like regions from multi-feature maps of different hierarchies and scales. Because the multi-feature maps combine the advantage of the deep and shallow convolutional layer, the first network performs well on locating the small targets in aerial image data. Then, the generated candidate regions are fed into the second network for feature extraction and decision making. Comprehensive experiments are conducted on the Vehicle Detection in Aerial Imagery (VEDAI) dataset and Munich vehicle dataset. The proposed cascaded detection model yields high performance, not only in detection accuracy but also in detection speed.

Non-linear feedback neural networks VLSI implementations and applications

CERN Document Server

Ansari, Mohd Samar

2014-01-01

This book aims to present a viable alternative to the Hopfield Neural Network (HNN) model for analog computation. It is well known that the standard HNN suffers from problems of convergence to local minima, and requirement of a large number of neurons and synaptic weights. Therefore, improved solutions are needed. The non-linear synapse neural network (NoSyNN) is one such possibility and is discussed in detail in this book. This book also discusses the applications in computationally intensive tasks like graph coloring, ranking, and linear as well as quadratic programming. The material in the book is useful to students, researchers and academician working in the area of analog computation.

Nonlinear adaptive PID control for greenhouse environment based on RBF network.

Science.gov (United States)

Zeng, Songwei; Hu, Haigen; Xu, Lihong; Li, Guanghui

2012-01-01

This paper presents a hybrid control strategy, combining Radial Basis Function (RBF) network with conventional proportional, integral, and derivative (PID) controllers, for the greenhouse climate control. A model of nonlinear conservation laws of enthalpy and matter between numerous system variables affecting the greenhouse climate is formulated. RBF network is used to tune and identify all PID gain parameters online and adaptively. The presented Neuro-PID control scheme is validated through simulations of set-point tracking and disturbance rejection. We compare the proposed adaptive online tuning method with the offline tuning scheme that employs Genetic Algorithm (GA) to search the optimal gain parameters. The results show that the proposed strategy has good adaptability, strong robustness and real-time performance while achieving satisfactory control performance for the complex and nonlinear greenhouse climate control system, and it may provide a valuable reference to formulate environmental control strategies for actual application in greenhouse production.

Absolute stability of nonlinear systems with time delays and applications to neural networks

Directory of Open Access Journals (Sweden)

Xinzhi Liu

2001-01-01

Full Text Available In this paper, absolute stability of nonlinear systems with time delays is investigated. Sufficient conditions on absolute stability are derived by using the comparison principle and differential inequalities. These conditions are simple and easy to check. In addition, exponential stability conditions for some special cases of nonlinear delay systems are discussed. Applications of those results to cellular neural networks are presented.

An Automatic Diagnosis Method of Facial Acne Vulgaris Based on Convolutional Neural Network.

Science.gov (United States)

Shen, Xiaolei; Zhang, Jiachi; Yan, Chenjun; Zhou, Hong

2018-04-11

In this paper, we present a new automatic diagnosis method for facial acne vulgaris which is based on convolutional neural networks (CNNs). To overcome the shortcomings of previous methods which were the inability to classify enough types of acne vulgaris. The core of our method is to extract features of images based on CNNs and achieve classification by classifier. A binary-classifier of skin-and-non-skin is used to detect skin area and a seven-classifier is used to achieve the classification task of facial acne vulgaris and healthy skin. In the experiments, we compare the effectiveness of our CNN and the VGG16 neural network which is pre-trained on the ImageNet data set. We use a ROC curve to evaluate the performance of binary-classifier and use a normalized confusion matrix to evaluate the performance of seven-classifier. The results of our experiments show that the pre-trained VGG16 neural network is effective in extracting features from facial acne vulgaris images. And the features are very useful for the follow-up classifiers. Finally, we try applying the classifiers both based on the pre-trained VGG16 neural network to assist doctors in facial acne vulgaris diagnosis.

Structural Observability and Sensor Node Selection for Complex Networks Governed by Nonlinear Balance Equations

NARCIS (Netherlands)

Kawano, Yu; Cao, Ming

2017-01-01

We define and then study the structural observability for a class of complex networks whose dynamics are governed by the nonlinear balance equations. Although related notions of observability of such complex networks have been studied before and in particular, necessary conditions have been reported

Nonlinear microrheology and molecular imaging to map microscale deformations of entangled DNA networks

Science.gov (United States)

Wu, Tsai-Chin; Anderson, Rae

We use active microrheology coupled to single-molecule fluorescence imaging to elucidate the microscale dynamics of entangled DNA. DNA naturally exists in a wide range of lengths and topologies, and is often confined in cell nucleui, forming highly concentrated and entangled biopolymer networks. Thus, DNA is the model polymer for understanding entangled polymer dynamics as well as the crowded environment of cells. These networks display complex viscoelastic properties that are not well understood, especially at the molecular-level and in response to nonlinear perturbations. Specifically, how microscopic stresses and strains propagate through entangled networks, and what molecular deformations lead to the network stress responses are unknown. To answer these important questions, we optically drive a microsphere through entangled DNA, perturbing the system far from equilibrium, while measuring the resistive force the DNA exerts on the bead during and after bead motion. We simultaneously image single fluorescent-labeled DNA molecules throughout the network to directly link the microscale stress response to molecular deformations. We characterize the deformation of the network from the molecular-level to the mesoscale, and map the stress propagation throughout the network. We further study the impact of DNA length (11 - 115 kbp) and topology (linear vs ring DNA) on deformation and propagation dynamics, exploring key nonlinear features such as tube dilation and power-law relaxation.

Pinning Synchronization of Delayed Neural Networks with Nonlinear Inner-Coupling

Directory of Open Access Journals (Sweden)

Yangling Wang

2011-01-01

Full Text Available Without assuming the symmetry and irreducibility of the outer-coupling weight configuration matrices, we investigate the pinning synchronization of delayed neural networks with nonlinear inner-coupling. Some delay-dependent controlled stability criteria in terms of linear matrix inequality (LMI are obtained. An example is presented to show the application of the criteria obtained in this paper.

Dynamic neural network-based methods for compensation of nonlinear effects in multimode communication lines

Science.gov (United States)

Sidelnikov, O. S.; Redyuk, A. A.; Sygletos, S.

2017-12-01

We consider neural network-based schemes of digital signal processing. It is shown that the use of a dynamic neural network-based scheme of signal processing ensures an increase in the optical signal transmission quality in comparison with that provided by other methods for nonlinear distortion compensation.

Convolutional neural network architectures for predicting DNA–protein binding

Science.gov (United States)

Zeng, Haoyang; Edwards, Matthew D.; Liu, Ge; Gifford, David K.

2016-01-01

Motivation: Convolutional neural networks (CNN) have outperformed conventional methods in modeling the sequence specificity of DNA–protein binding. Yet inappropriate CNN architectures can yield poorer performance than simpler models. Thus an in-depth understanding of how to match CNN architecture to a given task is needed to fully harness the power of CNNs for computational biology applications. Results: We present a systematic exploration of CNN architectures for predicting DNA sequence binding using a large compendium of transcription factor datasets. We identify the best-performing architectures by varying CNN width, depth and pooling designs. We find that adding convolutional kernels to a network is important for motif-based tasks. We show the benefits of CNNs in learning rich higher-order sequence features, such as secondary motifs and local sequence context, by comparing network performance on multiple modeling tasks ranging in difficulty. We also demonstrate how careful construction of sequence benchmark datasets, using approaches that control potentially confounding effects like positional or motif strength bias, is critical in making fair comparisons between competing methods. We explore how to establish the sufficiency of training data for these learning tasks, and we have created a flexible cloud-based framework that permits the rapid exploration of alternative neural network architectures for problems in computational biology. Availability and Implementation: All the models analyzed are available at http://cnn.csail.mit.edu. Contact: gifford@mit.edu Supplementary information: Supplementary data are available at Bioinformatics online. PMID:27307608

Cellular neural network to the spherical harmonics approximation of neutron transport equation in x–y geometry

International Nuclear Information System (INIS)

Pirouzmand, Ahmad; Hadad, Kamal

2012-01-01

Highlights: ► This paper describes the solution of time-dependent neutron transport equation. ► We use a novel method based on cellular neural networks (CNNs) coupled with the spherical harmonics method. ► We apply the CNN model to simulate step and ramp perturbation transients in a core. ► The accuracy and capabilities of the CNN model are examined for x–y geometry. - Abstract: In an earlier paper we utilized a novel method using cellular neural networks (CNNs) coupled with spherical harmonics method to solve the steady state neutron transport equation in x–y geometry. Here, the previous work is extended to the study of time-dependent neutron transport equation. To achieve this goal, an equivalent electrical circuit based on a second-order form of time-dependent neutron transport equation and one equivalent group of neutron precursor density is obtained by the CNN method. The CNN model is used to simulate step and ramp perturbation transients in a typical 2D core.

Advanced models of neural networks nonlinear dynamics and stochasticity in biological neurons

CERN Document Server

Rigatos, Gerasimos G

2015-01-01

This book provides a complete study on neural structures exhibiting nonlinear and stochastic dynamics, elaborating on neural dynamics by introducing advanced models of neural networks. It overviews the main findings in the modelling of neural dynamics in terms of electrical circuits and examines their stability properties with the use of dynamical systems theory. It is suitable for researchers and postgraduate students engaged with neural networks and dynamical systems theory.

Non-Linear Behaviour Of Gelatin Networks Reveals A Hierarchical Structure

KAUST Repository

Yang, Zhi; Hemar, Yacine; Hilliou, loic; Gilbert, Elliot P.; McGillivray, Duncan James; Williams, Martin A. K.; Chaieb, Saharoui

2015-01-01

We investigate the strain hardening behaviour of various gelatin networks - namely physically-crosslinked gelatin gel, chemically-crosslinked gelatin gels, and a hybrid gels made of a combination of the former two - under large shear deformations using the pre-stress, strain ramp, and large amplitude oscillation shear protocols. Further, the internal structures of physically-crosslinked gelatin gel and chemically-crosslinked gelatin gels were characterized by small angle neutron scattering (SANS) to enable their internal structures to be correlated with their nonlinear rheology. The Kratky plots of SANS data demonstrate the presence of small cross-linked aggregates within the chemically-crosslinked network, whereas in the physically-crosslinked gels a relatively homogeneous structure is observed. Through model fitting to the scattering data, we were able to obtain structural parameters, such as correlation length (ξ), cross-sectional polymer chain radius (Rc), and the fractal dimension (df) of the gel networks. The fractal dimension df obtained from the SANS data of the physically-crosslinked and chemically crosslinked gels is 1.31 and 1.53, respectively. These values are in excellent agreement with the ones obtained from a generalized non-linear elastic theory we used to fit our stress-strain curves. The chemical crosslinking that generates coils and aggregates hinders the free stretching of the triple helices bundles in the physically-crosslinked gels.

Non-Linear Behaviour Of Gelatin Networks Reveals A Hierarchical Structure

KAUST Repository

Yang, Zhi

2015-12-14

We investigate the strain hardening behaviour of various gelatin networks - namely physically-crosslinked gelatin gel, chemically-crosslinked gelatin gels, and a hybrid gels made of a combination of the former two - under large shear deformations using the pre-stress, strain ramp, and large amplitude oscillation shear protocols. Further, the internal structures of physically-crosslinked gelatin gel and chemically-crosslinked gelatin gels were characterized by small angle neutron scattering (SANS) to enable their internal structures to be correlated with their nonlinear rheology. The Kratky plots of SANS data demonstrate the presence of small cross-linked aggregates within the chemically-crosslinked network, whereas in the physically-crosslinked gels a relatively homogeneous structure is observed. Through model fitting to the scattering data, we were able to obtain structural parameters, such as correlation length (ξ), cross-sectional polymer chain radius (Rc), and the fractal dimension (df) of the gel networks. The fractal dimension df obtained from the SANS data of the physically-crosslinked and chemically crosslinked gels is 1.31 and 1.53, respectively. These values are in excellent agreement with the ones obtained from a generalized non-linear elastic theory we used to fit our stress-strain curves. The chemical crosslinking that generates coils and aggregates hinders the free stretching of the triple helices bundles in the physically-crosslinked gels.

Recurrence Density Enhanced Complex Networks for Nonlinear Time Series Analysis

Science.gov (United States)

Costa, Diego G. De B.; Reis, Barbara M. Da F.; Zou, Yong; Quiles, Marcos G.; Macau, Elbert E. N.

We introduce a new method, which is entitled Recurrence Density Enhanced Complex Network (RDE-CN), to properly analyze nonlinear time series. Our method first transforms a recurrence plot into a figure of a reduced number of points yet preserving the main and fundamental recurrence properties of the original plot. This resulting figure is then reinterpreted as a complex network, which is further characterized by network statistical measures. We illustrate the computational power of RDE-CN approach by time series by both the logistic map and experimental fluid flows, which show that our method distinguishes different dynamics sufficiently well as the traditional recurrence analysis. Therefore, the proposed methodology characterizes the recurrence matrix adequately, while using a reduced set of points from the original recurrence plots.

Forecasting hourly global solar radiation using hybrid k-means and nonlinear autoregressive neural network models

International Nuclear Information System (INIS)

Benmouiza, Khalil; Cheknane, Ali

2013-01-01

Highlights: • An unsupervised clustering algorithm with a neural network model was explored. • The forecasting results of solar radiation time series and the comparison of their performance was simulated. • A new method was proposed combining k-means algorithm and NAR network to provide better prediction results. - Abstract: In this paper, we review our work for forecasting hourly global horizontal solar radiation based on the combination of unsupervised k-means clustering algorithm and artificial neural networks (ANN). k-Means algorithm focused on extracting useful information from the data with the aim of modeling the time series behavior and find patterns of the input space by clustering the data. On the other hand, nonlinear autoregressive (NAR) neural networks are powerful computational models for modeling and forecasting nonlinear time series. Taking the advantage of both methods, a new method was proposed combining k-means algorithm and NAR network to provide better forecasting results

A deep convolutional neural network for recognizing foods

Science.gov (United States)

Jahani Heravi, Elnaz; Habibi Aghdam, Hamed; Puig, Domenec

2015-12-01

Controlling the food intake is an efficient way that each person can undertake to tackle the obesity problem in countries worldwide. This is achievable by developing a smartphone application that is able to recognize foods and compute their calories. State-of-art methods are chiefly based on hand-crafted feature extraction methods such as HOG and Gabor. Recent advances in large-scale object recognition datasets such as ImageNet have revealed that deep Convolutional Neural Networks (CNN) possess more representation power than the hand-crafted features. The main challenge with CNNs is to find the appropriate architecture for each problem. In this paper, we propose a deep CNN which consists of 769; 988 parameters. Our experiments show that the proposed CNN outperforms the state-of-art methods and improves the best result of traditional methods 17%. Moreover, using an ensemble of two CNNs that have been trained two different times, we are able to improve the classification performance 21:5%.

Deep learning with convolutional neural network in radiology.

Science.gov (United States)

Yasaka, Koichiro; Akai, Hiroyuki; Kunimatsu, Akira; Kiryu, Shigeru; Abe, Osamu

2018-04-01

Deep learning with a convolutional neural network (CNN) is gaining attention recently for its high performance in image recognition. Images themselves can be utilized in a learning process with this technique, and feature extraction in advance of the learning process is not required. Important features can be automatically learned. Thanks to the development of hardware and software in addition to techniques regarding deep learning, application of this technique to radiological images for predicting clinically useful information, such as the detection and the evaluation of lesions, etc., are beginning to be investigated. This article illustrates basic technical knowledge regarding deep learning with CNNs along the actual course (collecting data, implementing CNNs, and training and testing phases). Pitfalls regarding this technique and how to manage them are also illustrated. We also described some advanced topics of deep learning, results of recent clinical studies, and the future directions of clinical application of deep learning techniques.

Neural network based adaptive control for nonlinear dynamic regimes

Science.gov (United States)

Shin, Yoonghyun

Adaptive control designs using neural networks (NNs) based on dynamic inversion are investigated for aerospace vehicles which are operated at highly nonlinear dynamic regimes. NNs play a key role as the principal element of adaptation to approximately cancel the effect of inversion error, which subsequently improves robustness to parametric uncertainty and unmodeled dynamics in nonlinear regimes. An adaptive control scheme previously named 'composite model reference adaptive control' is further developed so that it can be applied to multi-input multi-output output feedback dynamic inversion. It can have adaptive elements in both the dynamic compensator (linear controller) part and/or in the conventional adaptive controller part, also utilizing state estimation information for NN adaptation. This methodology has more flexibility and thus hopefully greater potential than conventional adaptive designs for adaptive flight control in highly nonlinear flight regimes. The stability of the control system is proved through Lyapunov theorems, and validated with simulations. The control designs in this thesis also include the use of 'pseudo-control hedging' techniques which are introduced to prevent the NNs from attempting to adapt to various actuation nonlinearities such as actuator position and rate saturations. Control allocation is introduced for the case of redundant control effectors including thrust vectoring nozzles. A thorough comparison study of conventional and NN-based adaptive designs for a system under a limit cycle, wing-rock, is included in this research, and the NN-based adaptive control designs demonstrate their performances for two highly maneuverable aerial vehicles, NASA F-15 ACTIVE and FQM-117B unmanned aerial vehicle (UAV), operated under various nonlinearities and uncertainties.

Passivation and control of partially known SISO nonlinear systems via dynamic neural networks

Directory of Open Access Journals (Sweden)

Reyes-Reyes J.

2000-01-01

Full Text Available In this paper, an adaptive technique is suggested to provide the passivity property for a class of partially known SISO nonlinear systems. A simple Dynamic Neural Network (DNN, containing only two neurons and without any hidden-layers, is used to identify the unknown nonlinear system. By means of a Lyapunov-like analysis the new learning law for this DNN, guarantying both successful identification and passivation effects, is derived. Based on this adaptive DNN model, an adaptive feedback controller, serving for wide class of nonlinear systems with an a priori incomplete model description, is designed. Two typical examples illustrate the effectiveness of the suggested approach.

Nonlinear Silicon Photonic Signal Processing Devices for Future Optical Networks

Directory of Open Access Journals (Sweden)

Cosimo Lacava

2017-01-01

Full Text Available In this paper, we present a review on silicon-based nonlinear devices for all optical nonlinear processing of complex telecommunication signals. We discuss some recent developments achieved by our research group, through extensive collaborations with academic partners across Europe, on optical signal processing using silicon-germanium and amorphous silicon based waveguides as well as novel materials such as silicon rich silicon nitride and tantalum pentoxide. We review the performance of four wave mixing wavelength conversion applied on complex signals such as Differential Phase Shift Keying (DPSK, Quadrature Phase Shift Keying (QPSK, 16-Quadrature Amplitude Modulation (QAM and 64-QAM that dramatically enhance the telecom signal spectral efficiency, paving the way to next generation terabit all-optical networks.

An Application of Non-Linear Autoregressive Neural Networks to Predict Energy Consumption in Public Buildings

Directory of Open Access Journals (Sweden)

Luis Gonzaga Baca Ruiz

2016-08-01

Full Text Available This paper addresses the problem of energy consumption prediction using neural networks over a set of public buildings. Since energy consumption in the public sector comprises a substantial share of overall consumption, the prediction of such consumption represents a decisive issue in the achievement of energy savings. In our experiments, we use the data provided by an energy consumption monitoring system in a compound of faculties and research centers at the University of Granada, and provide a methodology to predict future energy consumption using nonlinear autoregressive (NAR and the nonlinear autoregressive neural network with exogenous inputs (NARX, respectively. Results reveal that NAR and NARX neural networks are both suitable for performing energy consumption prediction, but also that exogenous data may help to improve the accuracy of predictions.

Convolutional neural networks for vibrational spectroscopic data analysis.

Science.gov (United States)

Acquarelli, Jacopo; van Laarhoven, Twan; Gerretzen, Jan; Tran, Thanh N; Buydens, Lutgarde M C; Marchiori, Elena

2017-02-15

In this work we show that convolutional neural networks (CNNs) can be efficiently used to classify vibrational spectroscopic data and identify important spectral regions. CNNs are the current state-of-the-art in image classification and speech recognition and can learn interpretable representations of the data. These characteristics make CNNs a good candidate for reducing the need for preprocessing and for highlighting important spectral regions, both of which are crucial steps in the analysis of vibrational spectroscopic data. Chemometric analysis of vibrational spectroscopic data often relies on preprocessing methods involving baseline correction, scatter correction and noise removal, which are applied to the spectra prior to model building. Preprocessing is a critical step because even in simple problems using 'reasonable' preprocessing methods may decrease the performance of the final model. We develop a new CNN based method and provide an accompanying publicly available software. It is based on a simple CNN architecture with a single convolutional layer (a so-called shallow CNN). Our method outperforms standard classification algorithms used in chemometrics (e.g. PLS) in terms of accuracy when applied to non-preprocessed test data (86% average accuracy compared to the 62% achieved by PLS), and it achieves better performance even on preprocessed test data (96% average accuracy compared to the 89% achieved by PLS). For interpretability purposes, our method includes a procedure for finding important spectral regions, thereby facilitating qualitative interpretation of results. Copyright © 2016 Elsevier B.V. All rights reserved.

Crowd density estimation based on convolutional neural networks with mixed pooling

Science.gov (United States)

Zhang, Li; Zheng, Hong; Zhang, Ying; Zhang, Dongming

2017-09-01

Crowd density estimation is an important topic in the fields of machine learning and video surveillance. Existing methods do not provide satisfactory classification accuracy; moreover, they have difficulty in adapting to complex scenes. Therefore, we propose a method based on convolutional neural networks (CNNs). The proposed method improves performance of crowd density estimation in two key ways. First, we propose a feature pooling method named mixed pooling to regularize the CNNs. It replaces deterministic pooling operations with a parameter that, by studying the algorithm, could combine the conventional max pooling with average pooling methods. Second, we present a classification strategy, in which an image is divided into two cells and respectively categorized. The proposed approach was evaluated on three datasets: two ground truth image sequences and the University of California, San Diego, anomaly detection dataset. The results demonstrate that the proposed approach performs more effectively and easily than other methods.

A unifying view of synchronization for data assimilation in complex nonlinear networks

Science.gov (United States)

Abarbanel, Henry D. I.; Shirman, Sasha; Breen, Daniel; Kadakia, Nirag; Rey, Daniel; Armstrong, Eve; Margoliash, Daniel

2017-12-01

Networks of nonlinear systems contain unknown parameters and dynamical degrees of freedom that may not be observable with existing instruments. From observable state variables, we want to estimate the connectivity of a model of such a network and determine the full state of the model at the termination of a temporal observation window during which measurements transfer information to a model of the network. The model state at the termination of a measurement window acts as an initial condition for predicting the future behavior of the network. This allows the validation (or invalidation) of the model as a representation of the dynamical processes producing the observations. Once the model has been tested against new data, it may be utilized as a predictor of responses to innovative stimuli or forcing. We describe a general framework for the tasks involved in the "inverse" problem of determining properties of a model built to represent measured output from physical, biological, or other processes when the measurements are noisy, the model has errors, and the state of the model is unknown when measurements begin. This framework is called statistical data assimilation and is the best one can do in estimating model properties through the use of the conditional probability distributions of the model state variables, conditioned on observations. There is a very broad arena of applications of the methods described. These include numerical weather prediction, properties of nonlinear electrical circuitry, and determining the biophysical properties of functional networks of neurons. Illustrative examples will be given of (1) estimating the connectivity among neurons with known dynamics in a network of unknown connectivity, and (2) estimating the biophysical properties of individual neurons in vitro taken from a functional network underlying vocalization in songbirds.

Deep convolutional neural networks for automatic coronary calcium scoring in a screening study with low-dose chest CT

Science.gov (United States)

Lessmann, Nikolas; Išgum, Ivana; Setio, Arnaud A. A.; de Vos, Bob D.; Ciompi, Francesco; de Jong, Pim A.; Oudkerk, Matthjis; Mali, Willem P. Th. M.; Viergever, Max A.; van Ginneken, Bram

2016-03-01

The amount of calcifications in the coronary arteries is a powerful and independent predictor of cardiovascular events and is used to identify subjects at high risk who might benefit from preventive treatment. Routine quantification of coronary calcium scores can complement screening programs using low-dose chest CT, such as lung cancer screening. We present a system for automatic coronary calcium scoring based on deep convolutional neural networks (CNNs). The system uses three independently trained CNNs to estimate a bounding box around the heart. In this region of interest, connected components above 130 HU are considered candidates for coronary artery calcifications. To separate them from other high intensity lesions, classification of all extracted voxels is performed by feeding two-dimensional 50 mm × 50 mm patches from three orthogonal planes into three concurrent CNNs. The networks consist of three convolutional layers and one fully-connected layer with 256 neurons. In the experiments, 1028 non-contrast-enhanced and non-ECG-triggered low-dose chest CT scans were used. The network was trained on 797 scans. In the remaining 231 test scans, the method detected on average 194.3 mm3 of 199.8 mm3 coronary calcifications per scan (sensitivity 97.2 %) with an average false-positive volume of 10.3 mm3 . Subjects were assigned to one of five standard cardiovascular risk categories based on the Agatston score. Accuracy of risk category assignment was 84.4 % with a linearly weighted κ of 0.89. The proposed system can perform automatic coronary artery calcium scoring to identify subjects undergoing low-dose chest CT screening who are at risk of cardiovascular events with high accuracy.

Predicting non-linear dynamics by stable local learning in a recurrent spiking neural network.

Science.gov (United States)

Gilra, Aditya; Gerstner, Wulfram

2017-11-27

The brain needs to predict how the body reacts to motor commands, but how a network of spiking neurons can learn non-linear body dynamics using local, online and stable learning rules is unclear. Here, we present a supervised learning scheme for the feedforward and recurrent connections in a network of heterogeneous spiking neurons. The error in the output is fed back through fixed random connections with a negative gain, causing the network to follow the desired dynamics. The rule for Feedback-based Online Local Learning Of Weights (FOLLOW) is local in the sense that weight changes depend on the presynaptic activity and the error signal projected onto the postsynaptic neuron. We provide examples of learning linear, non-linear and chaotic dynamics, as well as the dynamics of a two-link arm. Under reasonable approximations, we show, using the Lyapunov method, that FOLLOW learning is uniformly stable, with the error going to zero asymptotically.

Application of convolutional artificial neural networks to echocardiograms for differentiating congenital heart diseases in a pediatric population

Science.gov (United States)

Perrin, Douglas P.; Bueno, Alejandra; Rodriguez, Andrea; Marx, Gerald R.; del Nido, Pedro J.

2017-03-01

In this paper we describe a pilot study, where machine learning methods are used to differentiate between congenital heart diseases. Our approach was to apply convolutional neural networks (CNNs) to echocardiographic images from five different pediatric populations: normal, coarctation of the aorta (CoA), hypoplastic left heart syndrome (HLHS), transposition of the great arteries (TGA), and single ventricle (SV). We used a single network topology that was trained in a pairwise fashion in order to evaluate the potential to differentiate between patient populations. In total we used 59,151 echo frames drawn from 1,666 clinical sequences. Approximately 80% of the data was used for training, and the remainder for validation. Data was split at sequence boundaries to avoid having related images in the training and validation sets. While training was done with echo images/frames, evaluation was performed for both single frame discrimination as well as sequence discrimination (by majority voting). In total 10 networks were generated and evaluated. Unlike other domains where this network topology has been used, in ultrasound there is low visual variation between classes. This work shows the potential for CNNs to be applied to this low-variation domain of medical imaging for disease discrimination.

On structure-exploiting trust-region regularized nonlinear least squares algorithms for neural-network learning.

Science.gov (United States)

Mizutani, Eiji; Demmel, James W

2003-01-01

This paper briefly introduces our numerical linear algebra approaches for solving structured nonlinear least squares problems arising from 'multiple-output' neural-network (NN) models. Our algorithms feature trust-region regularization, and exploit sparsity of either the 'block-angular' residual Jacobian matrix or the 'block-arrow' Gauss-Newton Hessian (or Fisher information matrix in statistical sense) depending on problem scale so as to render a large class of NN-learning algorithms 'efficient' in both memory and operation costs. Using a relatively large real-world nonlinear regression application, we shall explain algorithmic strengths and weaknesses, analyzing simulation results obtained by both direct and iterative trust-region algorithms with two distinct NN models: 'multilayer perceptrons' (MLP) and 'complementary mixtures of MLP-experts' (or neuro-fuzzy modular networks).

Training and Validating a Deep Convolutional Neural Network for Computer-Aided Detection and Classification of Abnormalities on Frontal Chest Radiographs.

Science.gov (United States)

Cicero, Mark; Bilbily, Alexander; Colak, Errol; Dowdell, Tim; Gray, Bruce; Perampaladas, Kuhan; Barfett, Joseph

2017-05-01

Convolutional neural networks (CNNs) are a subtype of artificial neural network that have shown strong performance in computer vision tasks including image classification. To date, there has been limited application of CNNs to chest radiographs, the most frequently performed medical imaging study. We hypothesize CNNs can learn to classify frontal chest radiographs according to common findings from a sufficiently large data set. Our institution's research ethics board approved a single-center retrospective review of 35,038 adult posterior-anterior chest radiographs and final reports performed between 2005 and 2015 (56% men, average age of 56, patient type: 24% inpatient, 39% outpatient, 37% emergency department) with a waiver for informed consent. The GoogLeNet CNN was trained using 3 graphics processing units to automatically classify radiographs as normal (n = 11,702) or into 1 or more of cardiomegaly (n = 9240), consolidation (n = 6788), pleural effusion (n = 7786), pulmonary edema (n = 1286), or pneumothorax (n = 1299). The network's performance was evaluated using receiver operating curve analysis on a test set of 2443 radiographs with the criterion standard being board-certified radiologist interpretation. Using 256 × 256-pixel images as input, the network achieved an overall sensitivity and specificity of 91% with an area under the curve of 0.964 for classifying a study as normal (n = 1203). For the abnormal categories, the sensitivity, specificity, and area under the curve, respectively, were 91%, 91%, and 0.962 for pleural effusion (n = 782), 82%, 82%, and 0.868 for pulmonary edema (n = 356), 74%, 75%, and 0.850 for consolidation (n = 214), 81%, 80%, and 0.875 for cardiomegaly (n = 482), and 78%, 78%, and 0.861 for pneumothorax (n = 167). Current deep CNN architectures can be trained with modest-sized medical data sets to achieve clinically useful performance at detecting and excluding common pathology on chest radiographs.

A pre-trained convolutional neural network based method for thyroid nodule diagnosis.

Science.gov (United States)

Ma, Jinlian; Wu, Fa; Zhu, Jiang; Xu, Dong; Kong, Dexing

2017-01-01

In ultrasound images, most thyroid nodules are in heterogeneous appearances with various internal components and also have vague boundaries, so it is difficult for physicians to discriminate malignant thyroid nodules from benign ones. In this study, we propose a hybrid method for thyroid nodule diagnosis, which is a fusion of two pre-trained convolutional neural networks (CNNs) with different convolutional layers and fully-connected layers. Firstly, the two networks pre-trained with ImageNet database are separately trained. Secondly, we fuse feature maps learned by trained convolutional filters, pooling and normalization operations of the two CNNs. Finally, with the fused feature maps, a softmax classifier is used to diagnose thyroid nodules. The proposed method is validated on 15,000 ultrasound images collected from two local hospitals. Experiment results show that the proposed CNN based methods can accurately and effectively diagnose thyroid nodules. In addition, the fusion of the two CNN based models lead to significant performance improvement, with an accuracy of 83.02%±0.72%. These demonstrate the potential clinical applications of this method. Copyright © 2016 Elsevier B.V. All rights reserved.

Flow-pattern identification and nonlinear dynamics of gas-liquid two-phase flow in complex networks.

Science.gov (United States)

Gao, Zhongke; Jin, Ningde

2009-06-01

The identification of flow pattern is a basic and important issue in multiphase systems. Because of the complexity of phase interaction in gas-liquid two-phase flow, it is difficult to discern its flow pattern objectively. In this paper, we make a systematic study on the vertical upward gas-liquid two-phase flow using complex network. Three unique network construction methods are proposed to build three types of networks, i.e., flow pattern complex network (FPCN), fluid dynamic complex network (FDCN), and fluid structure complex network (FSCN). Through detecting the community structure of FPCN by the community-detection algorithm based on K -mean clustering, useful and interesting results are found which can be used for identifying five vertical upward gas-liquid two-phase flow patterns. To investigate the dynamic characteristics of gas-liquid two-phase flow, we construct 50 FDCNs under different flow conditions, and find that the power-law exponent and the network information entropy, which are sensitive to the flow pattern transition, can both characterize the nonlinear dynamics of gas-liquid two-phase flow. Furthermore, we construct FSCN and demonstrate how network statistic can be used to reveal the fluid structure of gas-liquid two-phase flow. In this paper, from a different perspective, we not only introduce complex network theory to the study of gas-liquid two-phase flow but also indicate that complex network may be a powerful tool for exploring nonlinear time series in practice.

Neural Network Based Finite-Time Stabilization for Discrete-Time Markov Jump Nonlinear Systems with Time Delays

Directory of Open Access Journals (Sweden)

Fei Chen

2013-01-01

Full Text Available This paper deals with the finite-time stabilization problem for discrete-time Markov jump nonlinear systems with time delays and norm-bounded exogenous disturbance. The nonlinearities in different jump modes are parameterized by neural networks. Subsequently, a linear difference inclusion state space representation for a class of neural networks is established. Based on this, sufficient conditions are derived in terms of linear matrix inequalities to guarantee stochastic finite-time boundedness and stochastic finite-time stabilization of the closed-loop system. A numerical example is illustrated to verify the efficiency of the proposed technique.

Color Constancy Using CNNs

OpenAIRE

Bianco, Simone; Cusano, Claudio; Schettini, Raimondo

2015-01-01

In this work we describe a Convolutional Neural Network (CNN) to accurately predict the scene illumination. Taking image patches as input, the CNN works in the spatial domain without using hand-crafted features that are employed by most previous methods. The network consists of one convolutional layer with max pooling, one fully connected layer and three output nodes. Within the network structure, feature learning and regression are integrated into one optimization process, which leads to a m...

Distributed Containment Control for Multiple Unknown Second-Order Nonlinear Systems With Application to Networked Lagrangian Systems.

Science.gov (United States)

Mei, Jie; Ren, Wei; Li, Bing; Ma, Guangfu

2015-09-01

In this paper, we consider the distributed containment control problem for multiagent systems with unknown nonlinear dynamics. More specifically, we focus on multiple second-order nonlinear systems and networked Lagrangian systems. We first study the distributed containment control problem for multiple second-order nonlinear systems with multiple dynamic leaders in the presence of unknown nonlinearities and external disturbances under a general directed graph that characterizes the interaction among the leaders and the followers. A distributed adaptive control algorithm with an adaptive gain design based on the approximation capability of neural networks is proposed. We present a necessary and sufficient condition on the directed graph such that the containment error can be reduced as small as desired. As a byproduct, the leaderless consensus problem is solved with asymptotical convergence. Because relative velocity measurements between neighbors are generally more difficult to obtain than relative position measurements, we then propose a distributed containment control algorithm without using neighbors' velocity information. A two-step Lyapunov-based method is used to study the convergence of the closed-loop system. Next, we apply the ideas to deal with the containment control problem for networked unknown Lagrangian systems under a general directed graph. All the proposed algorithms are distributed and can be implemented using only local measurements in the absence of communication. Finally, simulation examples are provided to show the effectiveness of the proposed control algorithms.

Neural network modeling of nonlinear systems based on Volterra series extension of a linear model

Science.gov (United States)

Soloway, Donald I.; Bialasiewicz, Jan T.

1992-01-01

A Volterra series approach was applied to the identification of nonlinear systems which are described by a neural network model. A procedure is outlined by which a mathematical model can be developed from experimental data obtained from the network structure. Applications of the results to the control of robotic systems are discussed.

Predicting haemodynamic networks using electrophysiology: The role of non-linear and cross-frequency interactions

Science.gov (United States)

Tewarie, P.; Bright, M.G.; Hillebrand, A.; Robson, S.E.; Gascoyne, L.E.; Morris, P.G.; Meier, J.; Van Mieghem, P.; Brookes, M.J.

2016-01-01

Understanding the electrophysiological basis of resting state networks (RSNs) in the human brain is a critical step towards elucidating how inter-areal connectivity supports healthy brain function. In recent years, the relationship between RSNs (typically measured using haemodynamic signals) and electrophysiology has been explored using functional Magnetic Resonance Imaging (fMRI) and magnetoencephalography (MEG). Significant progress has been made, with similar spatial structure observable in both modalities. However, there is a pressing need to understand this relationship beyond simple visual similarity of RSN patterns. Here, we introduce a mathematical model to predict fMRI-based RSNs using MEG. Our unique model, based upon a multivariate Taylor series, incorporates both phase and amplitude based MEG connectivity metrics, as well as linear and non-linear interactions within and between neural oscillations measured in multiple frequency bands. We show that including non-linear interactions, multiple frequency bands and cross-frequency terms significantly improves fMRI network prediction. This shows that fMRI connectivity is not only the result of direct electrophysiological connections, but is also driven by the overlap of connectivity profiles between separate regions. Our results indicate that a complete understanding of the electrophysiological basis of RSNs goes beyond simple frequency-specific analysis, and further exploration of non-linear and cross-frequency interactions will shed new light on distributed network connectivity, and its perturbation in pathology. PMID:26827811

Robust model predictive control of nonlinear systems with unmodeled dynamics and bounded uncertainties based on neural networks.

Science.gov (United States)

Yan, Zheng; Wang, Jun

2014-03-01

This paper presents a neural network approach to robust model predictive control (MPC) for constrained discrete-time nonlinear systems with unmodeled dynamics affected by bounded uncertainties. The exact nonlinear model of underlying process is not precisely known, but a partially known nominal model is available. This partially known nonlinear model is first decomposed to an affine term plus an unknown high-order term via Jacobian linearization. The linearization residue combined with unmodeled dynamics is then modeled using an extreme learning machine via supervised learning. The minimax methodology is exploited to deal with bounded uncertainties. The minimax optimization problem is reformulated as a convex minimization problem and is iteratively solved by a two-layer recurrent neural network. The proposed neurodynamic approach to nonlinear MPC improves the computational efficiency and sheds a light for real-time implementability of MPC technology. Simulation results are provided to substantiate the effectiveness and characteristics of the proposed approach.

Adaptive fuzzy wavelet network control of second order multi-agent systems with unknown nonlinear dynamics.

Science.gov (United States)

Taheri, Mehdi; Sheikholeslam, Farid; Najafi, Majddedin; Zekri, Maryam

2017-07-01

In this paper, consensus problem is considered for second order multi-agent systems with unknown nonlinear dynamics under undirected graphs. A novel distributed control strategy is suggested for leaderless systems based on adaptive fuzzy wavelet networks. Adaptive fuzzy wavelet networks are employed to compensate for the effect of unknown nonlinear dynamics. Moreover, the proposed method is developed for leader following systems and leader following systems with state time delays. Lyapunov functions are applied to prove uniformly ultimately bounded stability of closed loop systems and to obtain adaptive laws. Three simulation examples are presented to illustrate the effectiveness of the proposed control algorithms. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.

Fast non-linear extraction of plasma equilibrium parameters using a neural network mapping

International Nuclear Information System (INIS)

Lister, J.B.; Schnurrenberger, H.

1990-07-01

The shaping of non-circular plasmas requires a non-linear mapping between the measured diagnostic signals and selected equilibrium parameters. The particular configuration of Neural Network known as the multi-layer perceptron provides a powerful and general technique for formulating an arbitrary continuous non-linear multi-dimensional mapping. This technique has been successfully applied to the extraction of equilibrium parameters from measurements of single-null diverted plasmas in the DIII-D tokamak; the results are compared with a purely linear mapping. The method is promising, and hardware implementation is straightforward. (author) 15 refs., 7 figs

Accurate detection of hierarchical communities in complex networks based on nonlinear dynamical evolution

Science.gov (United States)

Zhuo, Zhao; Cai, Shi-Min; Tang, Ming; Lai, Ying-Cheng

2018-04-01

One of the most challenging problems in network science is to accurately detect communities at distinct hierarchical scales. Most existing methods are based on structural analysis and manipulation, which are NP-hard. We articulate an alternative, dynamical evolution-based approach to the problem. The basic principle is to computationally implement a nonlinear dynamical process on all nodes in the network with a general coupling scheme, creating a networked dynamical system. Under a proper system setting and with an adjustable control parameter, the community structure of the network would "come out" or emerge naturally from the dynamical evolution of the system. As the control parameter is systematically varied, the community hierarchies at different scales can be revealed. As a concrete example of this general principle, we exploit clustered synchronization as a dynamical mechanism through which the hierarchical community structure can be uncovered. In particular, for quite arbitrary choices of the nonlinear nodal dynamics and coupling scheme, decreasing the coupling parameter from the global synchronization regime, in which the dynamical states of all nodes are perfectly synchronized, can lead to a weaker type of synchronization organized as clusters. We demonstrate the existence of optimal choices of the coupling parameter for which the synchronization clusters encode accurate information about the hierarchical community structure of the network. We test and validate our method using a standard class of benchmark modular networks with two distinct hierarchies of communities and a number of empirical networks arising from the real world. Our method is computationally extremely efficient, eliminating completely the NP-hard difficulty associated with previous methods. The basic principle of exploiting dynamical evolution to uncover hidden community organizations at different scales represents a "game-change" type of approach to addressing the problem of community

Image Classification Using Biomimetic Pattern Recognition with Convolutional Neural Networks Features

Science.gov (United States)

Huo, Guanying

2017-01-01

As a typical deep-learning model, Convolutional Neural Networks (CNNs) can be exploited to automatically extract features from images using the hierarchical structure inspired by mammalian visual system. For image classification tasks, traditional CNN models employ the softmax function for classification. However, owing to the limited capacity of the softmax function, there are some shortcomings of traditional CNN models in image classification. To deal with this problem, a new method combining Biomimetic Pattern Recognition (BPR) with CNNs is proposed for image classification. BPR performs class recognition by a union of geometrical cover sets in a high-dimensional feature space and therefore can overcome some disadvantages of traditional pattern recognition. The proposed method is evaluated on three famous image classification benchmarks, that is, MNIST, AR, and CIFAR-10. The classification accuracies of the proposed method for the three datasets are 99.01%, 98.40%, and 87.11%, respectively, which are much higher in comparison with the other four methods in most cases. PMID:28316614

Modeling cognitive deficits following neurodegenerative diseases and traumatic brain injuries with deep convolutional neural networks.

Science.gov (United States)

Lusch, Bethany; Weholt, Jake; Maia, Pedro D; Kutz, J Nathan

2018-06-01

The accurate diagnosis and assessment of neurodegenerative disease and traumatic brain injuries (TBI) remain open challenges. Both cause cognitive and functional deficits due to focal axonal swellings (FAS), but it is difficult to deliver a prognosis due to our limited ability to assess damaged neurons at a cellular level in vivo. We simulate the effects of neurodegenerative disease and TBI using convolutional neural networks (CNNs) as our model of cognition. We utilize biophysically relevant statistical data on FAS to damage the connections in CNNs in a functionally relevant way. We incorporate energy constraints on the brain by pruning the CNNs to be less over-engineered. Qualitatively, we demonstrate that damage leads to human-like mistakes. Our experiments also provide quantitative assessments of how accuracy is affected by various types and levels of damage. The deficit resulting from a fixed amount of damage greatly depends on which connections are randomly injured, providing intuition for why it is difficult to predict impairments. There is a large degree of subjectivity when it comes to interpreting cognitive deficits from complex systems such as the human brain. However, we provide important insight and a quantitative framework for disorders in which FAS are implicated. Copyright © 2018 Elsevier Inc. All rights reserved.

Weighted Optimization-Based Distributed Kalman Filter for Nonlinear Target Tracking in Collaborative Sensor Networks.

Science.gov (United States)

Chen, Jie; Li, Jiahong; Yang, Shuanghua; Deng, Fang

2017-11-01

The identification of the nonlinearity and coupling is crucial in nonlinear target tracking problem in collaborative sensor networks. According to the adaptive Kalman filtering (KF) method, the nonlinearity and coupling can be regarded as the model noise covariance, and estimated by minimizing the innovation or residual errors of the states. However, the method requires large time window of data to achieve reliable covariance measurement, making it impractical for nonlinear systems which are rapidly changing. To deal with the problem, a weighted optimization-based distributed KF algorithm (WODKF) is proposed in this paper. The algorithm enlarges the data size of each sensor by the received measurements and state estimates from its connected sensors instead of the time window. A new cost function is set as the weighted sum of the bias and oscillation of the state to estimate the "best" estimate of the model noise covariance. The bias and oscillation of the state of each sensor are estimated by polynomial fitting a time window of state estimates and measurements of the sensor and its neighbors weighted by the measurement noise covariance. The best estimate of the model noise covariance is computed by minimizing the weighted cost function using the exhaustive method. The sensor selection method is in addition to the algorithm to decrease the computation load of the filter and increase the scalability of the sensor network. The existence, suboptimality and stability analysis of the algorithm are given. The local probability data association method is used in the proposed algorithm for the multitarget tracking case. The algorithm is demonstrated in simulations on tracking examples for a random signal, one nonlinear target, and four nonlinear targets. Results show the feasibility and superiority of WODKF against other filtering algorithms for a large class of systems.

Noise-Induced Modulation of the Relaxation Kinetics around a Non-Equilibrium Steady State of Non-Linear Chemical Reaction Networks

OpenAIRE

Ramaswamy, Rajesh; Sbalzarini, Ivo F; González-Segredo, Nélido

2011-01-01

Stochastic effects from correlated noise non-trivially modulate the kinetics of non-linear chemical reaction networks. This is especially important in systems where reactions are confined to small volumes and reactants are delivered in bursts. We characterise how the two noise sources confinement and burst modulate the relaxation kinetics of a non-linear reaction network around a non-equilibrium steady state. We find that the lifetimes of species change with burst input and confinement. Confi...

Nonlinear aeroacoustic characterization of Helmholtz resonators with a local-linear neuro-fuzzy network model

Science.gov (United States)

Förner, K.; Polifke, W.

2017-10-01

The nonlinear acoustic behavior of Helmholtz resonators is characterized by a data-based reduced-order model, which is obtained by a combination of high-resolution CFD simulation and system identification. It is shown that even in the nonlinear regime, a linear model is capable of describing the reflection behavior at a particular amplitude with quantitative accuracy. This observation motivates to choose a local-linear model structure for this study, which consists of a network of parallel linear submodels. A so-called fuzzy-neuron layer distributes the input signal over the linear submodels, depending on the root mean square of the particle velocity at the resonator surface. The resulting model structure is referred to as an local-linear neuro-fuzzy network. System identification techniques are used to estimate the free parameters of this model from training data. The training data are generated by CFD simulations of the resonator, with persistent acoustic excitation over a wide range of frequencies and sound pressure levels. The estimated nonlinear, reduced-order models show good agreement with CFD and experimental data over a wide range of amplitudes for several test cases.

3D Convolutional Neural Network for Automatic Detection of Lung Nodules in Chest CT.

Science.gov (United States)

Hamidian, Sardar; Sahiner, Berkman; Petrick, Nicholas; Pezeshk, Aria

2017-01-01

Deep convolutional neural networks (CNNs) form the backbone of many state-of-the-art computer vision systems for classification and segmentation of 2D images. The same principles and architectures can be extended to three dimensions to obtain 3D CNNs that are suitable for volumetric data such as CT scans. In this work, we train a 3D CNN for automatic detection of pulmonary nodules in chest CT images using volumes of interest extracted from the LIDC dataset. We then convert the 3D CNN which has a fixed field of view to a 3D fully convolutional network (FCN) which can generate the score map for the entire volume efficiently in a single pass. Compared to the sliding window approach for applying a CNN across the entire input volume, the FCN leads to a nearly 800-fold speed-up, and thereby fast generation of output scores for a single case. This screening FCN is used to generate difficult negative examples that are used to train a new discriminant CNN. The overall system consists of the screening FCN for fast generation of candidate regions of interest, followed by the discrimination CNN.

Nonlinear model of epidemic spreading in a complex social network.

Science.gov (United States)

Kosiński, Robert A; Grabowski, A

2007-10-01

The epidemic spreading in a human society is a complex process, which can be described on the basis of a nonlinear mathematical model. In such an approach the complex and hierarchical structure of social network (which has implications for the spreading of pathogens and can be treated as a complex network), can be taken into account. In our model each individual has one of the four permitted states: susceptible, infected, infective, unsusceptible or dead. This refers to the SEIR model used in epidemiology. The state of an individual changes in time, depending on the previous state and the interactions with other individuals. The description of the interpersonal contacts is based on the experimental observations of the social relations in the community. It includes spatial localization of the individuals and hierarchical structure of interpersonal interactions. Numerical simulations were performed for different types of epidemics, giving the progress of a spreading process and typical relationships (e.g. range of epidemic in time, the epidemic curve). The spreading process has a complex and spatially chaotic character. The time dependence of the number of infective individuals shows the nonlinear character of the spreading process. We investigate the influence of the preventive vaccinations on the spreading process. In particular, for a critical value of preventively vaccinated individuals the percolation threshold is observed and the epidemic is suppressed.

Fast non-linear extraction of plasma equilibrium parameters using a neural network mapping

International Nuclear Information System (INIS)

Lister, J.B.; Schnurrenberger, H.

1991-01-01

The shaping of non-circular plasmas requires a non-linear mapping between the measured diagnostic signals and selected equilibrium parameters. The particular configuration of neural network known as the multilayer perceptron provides a powerful and general technique for formulating an arbitrary continuous non-linear multi-dimensional mapping. This technique has been successfully applied to the extraction of equilibrium parameters from measurements of single-null diverted plasmas in the DIII-D tokamak; the results are compared with a purely linear mapping. The method is promising, and hardware implementation is straightforward. (author). 17 refs, 8 figs, 2 tab

Application of nonlinear rheology to assess the effect of secondary nanofiller on network structure of hybrid polymer nanocomposites

Science.gov (United States)

Kamkar, Milad; Aliabadian, Ehsan; Shayesteh Zeraati, Ali; Sundararaj, Uttandaraman

2018-02-01

Carbon nanotube (CNT)/polymer nanocomposites exhibit excellent electrical properties by forming a percolated network. Adding a secondary filler can significantly affect the CNTs' network, resulting in changing the electrical properties. In this work, we investigated the effect of adding manganese dioxide nanowires (MnO2NWs) as a secondary nanofiller on the CNTs' network structure inside a poly(vinylidene fluoride) (PVDF) matrix. Incorporating MnO2NWs to PVDF/CNT samples produced a better state of dispersion of CNTs, as corroborated by light microscopy and transmission electron microscopy. The steady shear and oscillatory shear flows were employed to obtain a better insight into the nanofiller structure and viscoelastic behavior of the nanocomposites. The transient response under steady shear flow revealed that the stress overshoot of hybrid nanocomposites (two-fillers), PVDF/CNT/MnO2NWs, increased dramatically in comparison to binary nanocomposites (single-filler), PVDF/CNT and PVDF/MnO2NWs. This can be attributed to microstructural changes. Large amplitude oscillatory shear characterization was also performed to further investigate the effect of the secondary nanofiller on the nonlinear viscoelastic behavior of the samples. The nonlinear rheological observations were explained using quantitative nonlinear parameters [strain-stiffening ratio (S) and shear-thickening ratio (T)] and Lissajous-Bowditch plots. Results indicated that a more rigid nanofiller network was formed for the hybrid nanocomposites due to the better dispersion state of CNTs and this led to a more nonlinear viscoelastic behavior.

Transferring Deep Convolutional Neural Networks for the Scene Classification of High-Resolution Remote Sensing Imagery

Directory of Open Access Journals (Sweden)

Fan Hu

2015-11-01

Full Text Available Learning efficient image representations is at the core of the scene classification task of remote sensing imagery. The existing methods for solving the scene classification task, based on either feature coding approaches with low-level hand-engineered features or unsupervised feature learning, can only generate mid-level image features with limited representative ability, which essentially prevents them from achieving better performance. Recently, the deep convolutional neural networks (CNNs, which are hierarchical architectures trained on large-scale datasets, have shown astounding performance in object recognition and detection. However, it is still not clear how to use these deep convolutional neural networks for high-resolution remote sensing (HRRS scene classification. In this paper, we investigate how to transfer features from these successfully pre-trained CNNs for HRRS scene classification. We propose two scenarios for generating image features via extracting CNN features from different layers. In the first scenario, the activation vectors extracted from fully-connected layers are regarded as the final image features; in the second scenario, we extract dense features from the last convolutional layer at multiple scales and then encode the dense features into global image features through commonly used feature coding approaches. Extensive experiments on two public scene classification datasets demonstrate that the image features obtained by the two proposed scenarios, even with a simple linear classifier, can result in remarkable performance and improve the state-of-the-art by a significant margin. The results reveal that the features from pre-trained CNNs generalize well to HRRS datasets and are more expressive than the low- and mid-level features. Moreover, we tentatively combine features extracted from different CNN models for better performance.

Research on Adaptive Neural Network Control System Based on Nonlinear U-Model with Time-Varying Delay

Directory of Open Access Journals (Sweden)

Fengxia Xu

2014-01-01

Full Text Available U-model can approximate a large class of smooth nonlinear time-varying delay system to any accuracy by using time-varying delay parameters polynomial. This paper proposes a new approach, namely, U-model approach, to solving the problems of analysis and synthesis for nonlinear systems. Based on the idea of discrete-time U-model with time-varying delay, the identification algorithm of adaptive neural network is given for the nonlinear model. Then, the controller is designed by using the Newton-Raphson formula and the stability analysis is given for the closed-loop nonlinear systems. Finally, illustrative examples are given to show the validity and applicability of the obtained results.

Synchronization of chaotic recurrent neural networks with time-varying delays using nonlinear feedback control

International Nuclear Information System (INIS)

Cui Baotong; Lou Xuyang

2009-01-01

In this paper, a new method to synchronize two identical chaotic recurrent neural networks is proposed. Using the drive-response concept, a nonlinear feedback control law is derived to achieve the state synchronization of the two identical chaotic neural networks. Furthermore, based on the Lyapunov method, a delay independent sufficient synchronization condition in terms of linear matrix inequality (LMI) is obtained. A numerical example with graphical illustrations is given to illuminate the presented synchronization scheme

Neural network-based nonlinear model predictive control vs. linear quadratic gaussian control

Science.gov (United States)

Cho, C.; Vance, R.; Mardi, N.; Qian, Z.; Prisbrey, K.

1997-01-01

One problem with the application of neural networks to the multivariable control of mineral and extractive processes is determining whether and how to use them. The objective of this investigation was to compare neural network control to more conventional strategies and to determine if there are any advantages in using neural network control in terms of set-point tracking, rise time, settling time, disturbance rejection and other criteria. The procedure involved developing neural network controllers using both historical plant data and simulation models. Various control patterns were tried, including both inverse and direct neural network plant models. These were compared to state space controllers that are, by nature, linear. For grinding and leaching circuits, a nonlinear neural network-based model predictive control strategy was superior to a state space-based linear quadratic gaussian controller. The investigation pointed out the importance of incorporating state space into neural networks by making them recurrent, i.e., feeding certain output state variables into input nodes in the neural network. It was concluded that neural network controllers can have better disturbance rejection, set-point tracking, rise time, settling time and lower set-point overshoot, and it was also concluded that neural network controllers can be more reliable and easy to implement in complex, multivariable plants.

A Design Methodology for Efficient Implementation of Deconvolutional Neural Networks on an FPGA

OpenAIRE

Zhang, Xinyu; Das, Srinjoy; Neopane, Ojash; Kreutz-Delgado, Ken

2017-01-01

In recent years deep learning algorithms have shown extremely high performance on machine learning tasks such as image classification and speech recognition. In support of such applications, various FPGA accelerator architectures have been proposed for convolutional neural networks (CNNs) that enable high performance for classification tasks at lower power than CPU and GPU processors. However, to date, there has been little research on the use of FPGA implementations of deconvolutional neural...

[Computer aided diagnosis model for lung tumor based on ensemble convolutional neural network].

Science.gov (United States)

Wang, Yuanyuan; Zhou, Tao; Lu, Huiling; Wu, Cuiying; Yang, Pengfei

2017-08-01

The convolutional neural network (CNN) could be used on computer-aided diagnosis of lung tumor with positron emission tomography (PET)/computed tomography (CT), which can provide accurate quantitative analysis to compensate for visual inertia and defects in gray-scale sensitivity, and help doctors diagnose accurately. Firstly, parameter migration method is used to build three CNNs (CT-CNN, PET-CNN, and PET/CT-CNN) for lung tumor recognition in CT, PET, and PET/CT image, respectively. Then, we aimed at CT-CNN to obtain the appropriate model parameters for CNN training through analysis the influence of model parameters such as epochs, batchsize and image scale on recognition rate and training time. Finally, three single CNNs are used to construct ensemble CNN, and then lung tumor PET/CT recognition was completed through relative majority vote method and the performance between ensemble CNN and single CNN was compared. The experiment results show that the ensemble CNN is better than single CNN on computer-aided diagnosis of lung tumor.

Dynamical processes and epidemic threshold on nonlinear coupled multiplex networks

Science.gov (United States)

Gao, Chao; Tang, Shaoting; Li, Weihua; Yang, Yaqian; Zheng, Zhiming

2018-04-01

Recently, the interplay between epidemic spreading and awareness diffusion has aroused the interest of many researchers, who have studied models mainly based on linear coupling relations between information and epidemic layers. However, in real-world networks the relation between two layers may be closely correlated with the property of individual nodes and exhibits nonlinear dynamical features. Here we propose a nonlinear coupled information-epidemic model (I-E model) and present a comprehensive analysis in a more generalized scenario where the upload rate differs from node to node, deletion rate varies between susceptible and infected states, and infection rate changes between unaware and aware states. In particular, we develop a theoretical framework of the intra- and inter-layer dynamical processes with a microscopic Markov chain approach (MMCA), and derive an analytic epidemic threshold. Our results suggest that the change of upload and deletion rate has little effect on the diffusion dynamics in the epidemic layer.

Nonlinear analysis of gas-water/oil-water two-phase flow in complex networks

CERN Document Server

Gao, Zhong-Ke; Wang, Wen-Xu

2014-01-01

Understanding the dynamics of multi-phase flows has been a challenge in the fields of nonlinear dynamics and fluid mechanics. This chapter reviews our work on two-phase flow dynamics in combination with complex network theory. We systematically carried out gas-water/oil-water two-phase flow experiments for measuring the time series of flow signals which is studied in terms of the mapping from time series to complex networks. Three network mapping methods were proposed for the analysis and identification of flow patterns, i.e. Flow Pattern Complex Network (FPCN), Fluid Dynamic Complex Network (FDCN) and Fluid Structure Complex Network (FSCN). Through detecting the community structure of FPCN based on K-means clustering, distinct flow patterns can be successfully distinguished and identified. A number of FDCN’s under different flow conditions were constructed in order to reveal the dynamical characteristics of two-phase flows. The FDCNs exhibit universal power-law degree distributions. The power-law exponent ...

A novel method to produce nonlinear empirical physical formulas for experimental nonlinear electro-optical responses of doped nematic liquid crystals: Feedforward neural network approach

Energy Technology Data Exchange (ETDEWEB)

Yildiz, Nihat, E-mail: nyildiz@cumhuriyet.edu.t [Cumhuriyet University, Faculty of Science and Literature, Department of Physics, 58140 Sivas (Turkey); San, Sait Eren; Okutan, Mustafa [Department of Physics, Gebze Institute of Technology, P.O. Box 141, Gebze 41400, Kocaeli (Turkey); Kaya, Hueseyin [Cumhuriyet University, Faculty of Science and Literature, Department of Physics, 58140 Sivas (Turkey)

2010-04-15

Among other significant obstacles, inherent nonlinearity in experimental physical response data poses severe difficulty in empirical physical formula (EPF) construction. In this paper, we applied a novel method (namely layered feedforward neural network (LFNN) approach) to produce explicit nonlinear EPFs for experimental nonlinear electro-optical responses of doped nematic liquid crystals (NLCs). Our motivation was that, as we showed in a previous theoretical work, an appropriate LFNN, due to its exceptional nonlinear function approximation capabilities, is highly relevant to EPF construction. Therefore, in this paper, we obtained excellently produced LFNN approximation functions as our desired EPFs for above-mentioned highly nonlinear response data of NLCs. In other words, by using suitable LFNNs, we successfully fitted the experimentally measured response and predicted the new (yet-to-be measured) response data. The experimental data (response versus input) were diffraction and dielectric properties versus bias voltage; and they were all taken from our previous experimental work. We conclude that in general, LFNN can be applied to construct various types of EPFs for the corresponding various nonlinear physical perturbation (thermal, electronic, molecular, electric, optical, etc.) data of doped NLCs.

A novel method to produce nonlinear empirical physical formulas for experimental nonlinear electro-optical responses of doped nematic liquid crystals: Feedforward neural network approach

International Nuclear Information System (INIS)

Yildiz, Nihat; San, Sait Eren; Okutan, Mustafa; Kaya, Hueseyin

2010-01-01

Among other significant obstacles, inherent nonlinearity in experimental physical response data poses severe difficulty in empirical physical formula (EPF) construction. In this paper, we applied a novel method (namely layered feedforward neural network (LFNN) approach) to produce explicit nonlinear EPFs for experimental nonlinear electro-optical responses of doped nematic liquid crystals (NLCs). Our motivation was that, as we showed in a previous theoretical work, an appropriate LFNN, due to its exceptional nonlinear function approximation capabilities, is highly relevant to EPF construction. Therefore, in this paper, we obtained excellently produced LFNN approximation functions as our desired EPFs for above-mentioned highly nonlinear response data of NLCs. In other words, by using suitable LFNNs, we successfully fitted the experimentally measured response and predicted the new (yet-to-be measured) response data. The experimental data (response versus input) were diffraction and dielectric properties versus bias voltage; and they were all taken from our previous experimental work. We conclude that in general, LFNN can be applied to construct various types of EPFs for the corresponding various nonlinear physical perturbation (thermal, electronic, molecular, electric, optical, etc.) data of doped NLCs.

Distress Propagation in Complex Networks: The Case of Non-Linear DebtRank.

Directory of Open Access Journals (Sweden)

Marco Bardoscia

Full Text Available We consider a dynamical model of distress propagation on complex networks, which we apply to the study of financial contagion in networks of banks connected to each other by direct exposures. The model that we consider is an extension of the DebtRank algorithm, recently introduced in the literature. The mechanics of distress propagation is very simple: When a bank suffers a loss, distress propagates to its creditors, who in turn suffer losses, and so on. The original DebtRank assumes that losses are propagated linearly between connected banks. Here we relax this assumption and introduce a one-parameter family of non-linear propagation functions. As a case study, we apply this algorithm to a data-set of 183 European banks, and we study how the stability of the system depends on the non-linearity parameter under different stress-test scenarios. We find that the system is characterized by a transition between a regime where small shocks can be amplified and a regime where shocks do not propagate, and that the overall stability of the system increases between 2008 and 2013.

Diagonal recurrent neural network based adaptive control of nonlinear dynamical systems using lyapunov stability criterion.

Science.gov (United States)

Kumar, Rajesh; Srivastava, Smriti; Gupta, J R P

2017-03-01

In this paper adaptive control of nonlinear dynamical systems using diagonal recurrent neural network (DRNN) is proposed. The structure of DRNN is a modification of fully connected recurrent neural network (FCRNN). Presence of self-recurrent neurons in the hidden layer of DRNN gives it an ability to capture the dynamic behaviour of the nonlinear plant under consideration (to be controlled). To ensure stability, update rules are developed using lyapunov stability criterion. These rules are then used for adjusting the various parameters of DRNN. The responses of plants obtained with DRNN are compared with those obtained when multi-layer feed forward neural network (MLFFNN) is used as a controller. Also, in example 4, FCRNN is also investigated and compared with DRNN and MLFFNN. Robustness of the proposed control scheme is also tested against parameter variations and disturbance signals. Four simulation examples including one-link robotic manipulator and inverted pendulum are considered on which the proposed controller is applied. The results so obtained show the superiority of DRNN over MLFFNN as a controller. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.

Global Asymptotic Stability of Impulsive CNNs with Proportional Delays and Partially Lipschitz Activation Functions

Directory of Open Access Journals (Sweden)

Xueli Song

2014-01-01

Full Text Available This paper researches global asymptotic stability of impulsive cellular neural networks with proportional delays and partially Lipschitz activation functions. Firstly, by means of the transformation vi(t=ui(et, the impulsive cellular neural networks with proportional delays are transformed into impulsive cellular neural networks with the variable coefficients and constant delays. Secondly, we provide novel criteria for the uniqueness and exponential stability of the equilibrium point of the latter by relative nonlinear measure and prove that the exponential stability of equilibrium point of the latter implies the asymptotic stability of one of the former. We furthermore obtain a sufficient condition to the uniqueness and global asymptotic stability of the equilibrium point of the former. Our method does not require conventional assumptions on global Lipschitz continuity, boundedness, and monotonicity of activation functions. Our results are generalizations and improvements of some existing ones. Finally, an example and its simulations are provided to illustrate the correctness of our analysis.

Neural networks for tracking of unknown SISO discrete-time nonlinear dynamic systems.

Science.gov (United States)

Aftab, Muhammad Saleheen; Shafiq, Muhammad

2015-11-01

This article presents a Lyapunov function based neural network tracking (LNT) strategy for single-input, single-output (SISO) discrete-time nonlinear dynamic systems. The proposed LNT architecture is composed of two feedforward neural networks operating as controller and estimator. A Lyapunov function based back propagation learning algorithm is used for online adjustment of the controller and estimator parameters. The controller and estimator error convergence and closed-loop system stability analysis is performed by Lyapunov stability theory. Moreover, two simulation examples and one real-time experiment are investigated as case studies. The achieved results successfully validate the controller performance. Copyright © 2015 ISA. Published by Elsevier Ltd. All rights reserved.

Pre-Trained Neural Networks used for Non-Linear State Estimation

DEFF Research Database (Denmark)

Bayramoglu, Enis; Andersen, Nils Axel; Ravn, Ole

2011-01-01

of the paramters in the distribution. This transformation is approximated by a neural network using offline training, which is based on monte carlo sampling. In the paper, there will also be presented a method to construct a flexible distributions well suited for covering the effect of the non-linearities......The paper focuses on nonlinear state estimation assuming non-Gaussian distributions of the states and the disturbances. The posterior distribution and the aposteriori distribution is described by a chosen family of paramtric distributions. The state transformation then results in a transformation...

Reluctance network modelling of surface permanent magnet motor considering iron nonlinearities

International Nuclear Information System (INIS)

Raminosoa, T.; Farooq, J.A.; Djerdir, A.; Miraoui, A.

2009-01-01

This paper presents a simple, quick and precise nonlinear reluctance network modelling of an in-wheel surface permanent magnet motor. The saturation of the ferromagnetic materials is considered and a simple air-gap length function is used to take the slotting effect into account. The topology and the reluctance values of the air-gap network are automatically computed for any rotor position. Thus, the proposed technique allows a steady state time stepping simulation. For any saturation level, there is a good accordance with the finite element method for the torque and back EMF. Moreover, the model accurately predicts the effect of the demagnetization. The proposed model can be advantageously used for a geometry optimization as well as for the diagnosis of demagnetization.

Event-Triggered Fault Detection of Nonlinear Networked Systems.

Science.gov (United States)

Li, Hongyi; Chen, Ziran; Wu, Ligang; Lam, Hak-Keung; Du, Haiping

2017-04-01

This paper investigates the problem of fault detection for nonlinear discrete-time networked systems under an event-triggered scheme. A polynomial fuzzy fault detection filter is designed to generate a residual signal and detect faults in the system. A novel polynomial event-triggered scheme is proposed to determine the transmission of the signal. A fault detection filter is designed to guarantee that the residual system is asymptotically stable and satisfies the desired performance. Polynomial approximated membership functions obtained by Taylor series are employed for filtering analysis. Furthermore, sufficient conditions are represented in terms of sum of squares (SOSs) and can be solved by SOS tools in MATLAB environment. A numerical example is provided to demonstrate the effectiveness of the proposed results.

Linear modeling of nonlinear systems using artificial neural networks based on I/O data and its application in power plant boiler modeling

International Nuclear Information System (INIS)

Ghaffari, A.; Nikkhah Bahrami, M.; Mohammadzaheri, M.

2005-01-01

In this paper a new method for linear modeling of nonlinear systems is presented. The method is based on the design of an artificial neural network with two layers. The network is trained only according to the input-output data of the system. The weights of connections in this network, represents the coefficients of the transfer function. For systems with linear behavior the method of least square error represents the best linear model of the system. However, for nonlinear systems, such as some subsystems in power plants boilers LSE does not represent the best linear approximation of the system, necessarily. In this paper a new linear modeling method is presented and applied to some subsystems in a power plant boiler. Comparison between the transfer function obtained in this way and by least square error method,shows that the neural network method gives better linear models for these nonlinear systems

Nonlinear analysis and synthesis of video images using deep dynamic bottleneck neural networks for face recognition.

Science.gov (United States)

Moghadam, Saeed Montazeri; Seyyedsalehi, Seyyed Ali

2018-05-31

Nonlinear components extracted from deep structures of bottleneck neural networks exhibit a great ability to express input space in a low-dimensional manifold. Sharing and combining the components boost the capability of the neural networks to synthesize and interpolate new and imaginary data. This synthesis is possibly a simple model of imaginations in human brain where the components are expressed in a nonlinear low dimensional manifold. The current paper introduces a novel Dynamic Deep Bottleneck Neural Network to analyze and extract three main features of videos regarding the expression of emotions on the face. These main features are identity, emotion and expression intensity that are laid in three different sub-manifolds of one nonlinear general manifold. The proposed model enjoying the advantages of recurrent networks was used to analyze the sequence and dynamics of information in videos. It is noteworthy to mention that this model also has also the potential to synthesize new videos showing variations of one specific emotion on the face of unknown subjects. Experiments on discrimination and recognition ability of extracted components showed that the proposed model has an average of 97.77% accuracy in recognition of six prominent emotions (Fear, Surprise, Sadness, Anger, Disgust, and Happiness), and 78.17% accuracy in the recognition of intensity. The produced videos revealed variations from neutral to the apex of an emotion on the face of the unfamiliar test subject which is on average 0.8 similar to reference videos in the scale of the SSIM method. Copyright © 2018 Elsevier Ltd. All rights reserved.

Can we recognize horses by their ocular biometric traits using deep convolutional neural networks?

Science.gov (United States)

Trokielewicz, Mateusz; Szadkowski, Mateusz

2017-08-01

This paper aims at determining the viability of horse recognition by the means of ocular biometrics and deep convolutional neural networks (deep CNNs). Fast and accurate identification of race horses before racing is crucial for ensuring that exactly the horses that were declared are participating, using methods that are non-invasive and friendly to these delicate animals. As typical iris recognition methods require lot of fine-tuning of the method parameters and high-quality data, CNNs seem like a natural candidate to be applied for recognition thanks to their potentially excellent abilities in describing texture, combined with ease of implementation in an end-to-end manner. Also, with such approach we can easily utilize both iris and periocular features without constructing complicated algorithms for each. We thus present a simple CNN classifier, able to correctly identify almost 80% of the samples in an identification scenario, and give equal error rate (EER) of less than 10% in a verification scenario.

Global stability and existence of periodic solutions of discrete delayed cellular neural networks

International Nuclear Information System (INIS)

Li Yongkun

2004-01-01

We use the continuation theorem of coincidence degree theory and Lyapunov functions to study the existence and stability of periodic solutions for the discrete cellular neural networks (CNNs) with delays xi(n+1)=xi(n)e-bi(n)h+θi(h)-bar j=1maij(n)fj(xj(n))+θi(h)-bar j=1mbij(n)fj(xj(n- τij(n)))+θi(h)Ii(n),i=1,2,...,m. We obtain some sufficient conditions to ensure that for the networks there exists a unique periodic solution, and all its solutions converge to such a periodic solution

A mixed-integer nonlinear programming approach to the optimal design of heat network in a polygeneration energy system

International Nuclear Information System (INIS)

Zhang, Jianyun; Liu, Pei; Zhou, Zhe; Ma, Linwei; Li, Zheng; Ni, Weidou

2014-01-01

Highlights: • Integration of heat streams with HRSG in a polygeneration system is studied. • A mixed-integer nonlinear programming model is proposed to optimize heat network. • Operating parameters and heat network configuration are optimized simultaneously. • The optimized heat network highly depends on the HRSG type and model specification. - Abstract: A large number of heat flows at various temperature and pressure levels exist in a polygeneration plant which co-produces electricity and chemical products. Integration of these external heat flows in a heat recovery steam generator (HRSG) has great potential to further enhance energy efficiency of such a plant; however, it is a challenging problem arising from the large design space of heat exchanger network. In this paper, a mixed-integer nonlinear programming model is developed for the design optimization of a HRSG with consideration of all alternative matches between the HRSG and external heat flows. This model is applied to four polygeneration cases with different HRSG types, and results indicate that the optimized heat network mainly depends on the HRSG type and the model specification

Transient response of nonlinear polymer networks: A kinetic theory

Science.gov (United States)

Vernerey, Franck J.

2018-06-01

Dynamic networks are found in a majority of natural materials, but also in engineering materials, such as entangled polymers and physically cross-linked gels. Owing to their transient bond dynamics, these networks display a rich class of behaviors, from elasticity, rheology, self-healing, or growth. Although classical theories in rheology and mechanics have enabled us to characterize these materials, there is still a gap in our understanding on how individuals (i.e., the mechanics of each building blocks and its connection with others) affect the emerging response of the network. In this work, we introduce an alternative way to think about these networks from a statistical point of view. More specifically, a network is seen as a collection of individual polymer chains connected by weak bonds that can associate and dissociate over time. From the knowledge of these individual chains (elasticity, transient attachment, and detachment events), we construct a statistical description of the population and derive an evolution equation of their distribution based on applied deformation and their local interactions. We specifically concentrate on nonlinear elastic response that follows from the strain stiffening response of individual chains of finite size. Upon appropriate averaging operations and using a mean field approximation, we show that the distribution can be replaced by a so-called chain distribution tensor that is used to determine important macroscopic measures such as stress, energy storage and dissipation in the network. Prediction of the kinetic theory are then explored against known experimental measurement of polymer responses under uniaxial loading. It is found that even under the simplest assumptions of force-independent chain kinetics, the model is able to reproduce complex time-dependent behaviors of rubber and self-healing supramolecular polymers.

Impulsive Controller Design for Complex Nonlinear Singular Networked Systems with Packet Dropouts

Directory of Open Access Journals (Sweden)

Xian-Lin Zhao

2013-01-01

Full Text Available Globally exponential stability of Complex (with coupling Nonlinear Singular Impulsive Networked Control Systems (CNSINCS with packet dropouts and time-delay is investigated. Firstly, the mathematic model of CNSINCS is established. Then, by employing the method of Lyapunov functional, exponential stability criteria are obtained and the impulsive controller design method is given. Finally, some simulation results are provided to demonstrate the effectiveness of the proposed method.

Analysis on evaluation ability of nonlinear safety assessment model of coal mines based on artificial neural network

Institute of Scientific and Technical Information of China (English)

SHI Shi-liang; LIU Hai-bo; LIU Ai-hua

2004-01-01

Based on the integration analysis of goods and shortcomings of various methods used in safety assessment of coal mines, combining nonlinear feature of mine safety sub-system, this paper establishes the neural network assessment model of mine safety, analyzes the ability of artificial neural network to evaluate mine safety state, and lays the theoretical foundation of artificial neural network using in the systematic optimization of mine safety assessment and getting reasonable accurate safety assessment result.

Nonlinear random resistor diode networks and fractal dimensions of directed percolation clusters.

Science.gov (United States)

Stenull, O; Janssen, H K

2001-07-01

We study nonlinear random resistor diode networks at the transition from the nonpercolating to the directed percolating phase. The resistor-like bonds and the diode-like bonds under forward bias voltage obey a generalized Ohm's law V approximately I(r). Based on general grounds such as symmetries and relevance we develop a field theoretic model. We focus on the average two-port resistance, which is governed at the transition by the resistance exponent straight phi(r). By employing renormalization group methods we calculate straight phi(r) for arbitrary r to one-loop order. Then we address the fractal dimensions characterizing directed percolation clusters. Via considering distinct values of the nonlinearity r, we determine the dimension of the red bonds, the chemical path, and the backbone to two-loop order.

Automated diagnosis of prostate cancer in multi-parametric MRI based on multimodal convolutional neural networks

Science.gov (United States)

Le, Minh Hung; Chen, Jingyu; Wang, Liang; Wang, Zhiwei; Liu, Wenyu; (Tim Cheng, Kwang-Ting; Yang, Xin

2017-08-01

Automated methods for prostate cancer (PCa) diagnosis in multi-parametric magnetic resonance imaging (MP-MRIs) are critical for alleviating requirements for interpretation of radiographs while helping to improve diagnostic accuracy (Artan et al 2010 IEEE Trans. Image Process. 19 2444-55, Litjens et al 2014 IEEE Trans. Med. Imaging 33 1083-92, Liu et al 2013 SPIE Medical Imaging (International Society for Optics and Photonics) p 86701G, Moradi et al 2012 J. Magn. Reson. Imaging 35 1403-13, Niaf et al 2014 IEEE Trans. Image Process. 23 979-91, Niaf et al 2012 Phys. Med. Biol. 57 3833, Peng et al 2013a SPIE Medical Imaging (International Society for Optics and Photonics) p 86701H, Peng et al 2013b Radiology 267 787-96, Wang et al 2014 BioMed. Res. Int. 2014). This paper presents an automated method based on multimodal convolutional neural networks (CNNs) for two PCa diagnostic tasks: (1) distinguishing between cancerous and noncancerous tissues and (2) distinguishing between clinically significant (CS) and indolent PCa. Specifically, our multimodal CNNs effectively fuse apparent diffusion coefficients (ADCs) and T2-weighted MP-MRI images (T2WIs). To effectively fuse ADCs and T2WIs we design a new similarity loss function to enforce consistent features being extracted from both ADCs and T2WIs. The similarity loss is combined with the conventional classification loss functions and integrated into the back-propagation procedure of CNN training. The similarity loss enables better fusion results than existing methods as the feature learning processes of both modalities are mutually guided, jointly facilitating CNN to ‘see’ the true visual patterns of PCa. The classification results of multimodal CNNs are further combined with the results based on handcrafted features using a support vector machine classifier. To achieve a satisfactory accuracy for clinical use, we comprehensively investigate three critical factors which could greatly affect the performance of our

Discriminating between benign and malignant breast tumors using 3D convolutional neural network in dynamic contrast enhanced-MR images

Science.gov (United States)

Li, Jing; Fan, Ming; Zhang, Juan; Li, Lihua

2017-03-01

Convolutional neural networks (CNNs) are the state-of-the-art deep learning network architectures that can be used in a range of applications, including computer vision and medical image analysis. It exhibits a powerful representation learning mechanism with an automated design to learn features directly from the data. However, the common 2D CNNs only use the two dimension spatial information without evaluating the correlation between the adjoin slices. In this study, we established a method of 3D CNNs to discriminate between malignant and benign breast tumors. To this end, 143 patients were enrolled which include 66 benign and 77 malignant instances. The MRI images were pre-processed for noise reduction and breast tumor region segmentation. Data augmentation by spatial translating, rotating and vertical and horizontal flipping is applied to the cases to reduce possible over-fitting. A region-of-interest (ROI) and a volume-of-interest (VOI) were segmented in 2D and 3D DCE-MRI, respectively. The enhancement ratio for each MR series was calculated for the 2D and 3D images. The results for the enhancement ratio images in the two series are integrated for classification. The results of the area under the ROC curve(AUC) values are 0.739 and 0.801 for 2D and 3D methods, respectively. The results for 3D CNN which combined 5 slices for each enhancement ratio images achieved a high accuracy(Acc), sensitivity(Sens) and specificity(Spec) of 0.781, 0.744 and 0.823, respectively. This study indicates that 3D CNN deep learning methods can be a promising technology for breast tumor classification without manual feature extraction.

Fuzzy Wavelet Neural Network Using a Correntropy Criterion for Nonlinear System Identification

Directory of Open Access Journals (Sweden)

Leandro L. S. Linhares

2015-01-01

Full Text Available Recent researches have demonstrated that the Fuzzy Wavelet Neural Networks (FWNNs are an efficient tool to identify nonlinear systems. In these structures, features related to fuzzy logic, wavelet functions, and neural networks are combined in an architecture similar to the Adaptive Neurofuzzy Inference Systems (ANFIS. In practical applications, the experimental data set used in the identification task often contains unknown noise and outliers, which decrease the FWNN model reliability. In order to reduce the negative effects of these erroneous measurements, this work proposes the direct use of a similarity measure based on information theory in the FWNN learning procedure. The Mean Squared Error (MSE cost function is replaced by the Maximum Correntropy Criterion (MCC in the traditional error backpropagation (BP algorithm. The input-output maps of a real nonlinear system studied in this work are identified from an experimental data set corrupted by different outliers rates and additive white Gaussian noise. The results demonstrate the advantages of the proposed cost function using the MCC as compared to the MSE. This work also investigates the influence of the kernel size on the performance of the MCC in the BP algorithm, since it is the only free parameter of correntropy.

Robust nonlinear autoregressive moving average model parameter estimation using stochastic recurrent artificial neural networks

DEFF Research Database (Denmark)

Chon, K H; Hoyer, D; Armoundas, A A

1999-01-01

In this study, we introduce a new approach for estimating linear and nonlinear stochastic autoregressive moving average (ARMA) model parameters, given a corrupt signal, using artificial recurrent neural networks. This new approach is a two-step approach in which the parameters of the deterministic...... part of the stochastic ARMA model are first estimated via a three-layer artificial neural network (deterministic estimation step) and then reestimated using the prediction error as one of the inputs to the artificial neural networks in an iterative algorithm (stochastic estimation step). The prediction...... error is obtained by subtracting the corrupt signal of the estimated ARMA model obtained via the deterministic estimation step from the system output response. We present computer simulation examples to show the efficacy of the proposed stochastic recurrent neural network approach in obtaining accurate...

Switching Fuzzy Guaranteed Cost Control for Nonlinear Networked Control Systems

Directory of Open Access Journals (Sweden)

Linqin Cai

2014-01-01

Full Text Available This paper deals with the problem of guaranteed cost control for a class of nonlinear networked control systems (NCSs with time-varying delay. A guaranteed cost controller design method is proposed to achieve the desired control performance based on the switched T-S fuzzy model. The switching mechanism is introduced to handle the uncertainties of NCSs. Based on Lyapunov functional approach, some sufficient conditions for the existence of state feedback robust guaranteed cost controller are presented. Simulation results show that the proposed method is effective to guarantee system’s global asymptotic stability and quality of service (QoS.

Neural Networks for the Classification of Building Use from Street-View Imagery

Science.gov (United States)

Laupheimer, D.; Tutzauer, P.; Haala, N.; Spicker, M.

2018-05-01

Within this paper we propose an end-to-end approach for classifying terrestrial images of building facades into five different utility classes (commercial, hybrid, residential, specialUse, underConstruction) by using Convolutional Neural Networks (CNNs). For our examples we use images provided by Google Street View. These images are automatically linked to a coarse city model, including the outlines of the buildings as well as their respective use classes. By these means an extensive dataset is available for training and evaluation of our Deep Learning pipeline. The paper describes the implemented end-to-end approach for classifying street-level images of building facades and discusses our experiments with various CNNs. In addition to the classification results, so-called Class Activation Maps (CAMs) are evaluated. These maps give further insights into decisive facade parts that are learned as features during the training process. Furthermore, they can be used for the generation of abstract presentations which facilitate the comprehension of semantic image content. The abstract representations are a result of the stippling method, an importance-based image rendering.

Fully automatic acute ischemic lesion segmentation in DWI using convolutional neural networks

Directory of Open Access Journals (Sweden)

Liang Chen

2017-01-01

Full Text Available Stroke is an acute cerebral vascular disease, which is likely to cause long-term disabilities and death. Acute ischemic lesions occur in most stroke patients. These lesions are treatable under accurate diagnosis and treatments. Although diffusion-weighted MR imaging (DWI is sensitive to these lesions, localizing and quantifying them manually is costly and challenging for clinicians. In this paper, we propose a novel framework to automatically segment stroke lesions in DWI. Our framework consists of two convolutional neural networks (CNNs: one is an ensemble of two DeconvNets (Noh et al., 2015, which is the EDD Net; the second CNN is the multi-scale convolutional label evaluation net (MUSCLE Net, which aims to evaluate the lesions detected by the EDD Net in order to remove potential false positives. To the best of our knowledge, it is the first attempt to solve this problem and using both CNNs achieves very good results. Furthermore, we study the network architectures and key configurations in detail to ensure the best performance. It is validated on a large dataset comprising clinical acquired DW images from 741 subjects. A mean accuracy of Dice coefficient obtained is 0.67 in total. The mean Dice scores based on subjects with only small and large lesions are 0.61 and 0.83, respectively. The lesion detection rate achieved is 0.94.

Fully automatic acute ischemic lesion segmentation in DWI using convolutional neural networks.

Science.gov (United States)

Chen, Liang; Bentley, Paul; Rueckert, Daniel

2017-01-01

Stroke is an acute cerebral vascular disease, which is likely to cause long-term disabilities and death. Acute ischemic lesions occur in most stroke patients. These lesions are treatable under accurate diagnosis and treatments. Although diffusion-weighted MR imaging (DWI) is sensitive to these lesions, localizing and quantifying them manually is costly and challenging for clinicians. In this paper, we propose a novel framework to automatically segment stroke lesions in DWI. Our framework consists of two convolutional neural networks (CNNs): one is an ensemble of two DeconvNets (Noh et al., 2015), which is the EDD Net; the second CNN is the multi-scale convolutional label evaluation net (MUSCLE Net), which aims to evaluate the lesions detected by the EDD Net in order to remove potential false positives. To the best of our knowledge, it is the first attempt to solve this problem and using both CNNs achieves very good results. Furthermore, we study the network architectures and key configurations in detail to ensure the best performance. It is validated on a large dataset comprising clinical acquired DW images from 741 subjects. A mean accuracy of Dice coefficient obtained is 0.67 in total. The mean Dice scores based on subjects with only small and large lesions are 0.61 and 0.83, respectively. The lesion detection rate achieved is 0.94.

Nonlinear Inference in Partially Observed Physical Systems and Deep Neural Networks

Science.gov (United States)

Rozdeba, Paul J.

The problem of model state and parameter estimation is a significant challenge in nonlinear systems. Due to practical considerations of experimental design, it is often the case that physical systems are partially observed, meaning that data is only available for a subset of the degrees of freedom required to fully model the observed system's behaviors and, ultimately, predict future observations. Estimation in this context is highly complicated by the presence of chaos, stochasticity, and measurement noise in dynamical systems. One of the aims of this dissertation is to simultaneously analyze state and parameter estimation in as a regularized inverse problem, where the introduction of a model makes it possible to reverse the forward problem of partial, noisy observation; and as a statistical inference problem using data assimilation to transfer information from measurements to the model states and parameters. Ultimately these two formulations achieve the same goal. Similar aspects that appear in both are highlighted as a means for better understanding the structure of the nonlinear inference problem. An alternative approach to data assimilation that uses model reduction is then examined as a way to eliminate unresolved nonlinear gating variables from neuron models. In this formulation, only measured variables enter into the model, and the resulting errors are themselves modeled by nonlinear stochastic processes with memory. Finally, variational annealing, a data assimilation method previously applied to dynamical systems, is introduced as a potentially useful tool for understanding deep neural network training in machine learning by exploiting similarities between the two problems.

Adaptive neural network output feedback control for stochastic nonlinear systems with unknown dead-zone and unmodeled dynamics.

Science.gov (United States)

Tong, Shaocheng; Wang, Tong; Li, Yongming; Zhang, Huaguang

2014-06-01

This paper discusses the problem of adaptive neural network output feedback control for a class of stochastic nonlinear strict-feedback systems. The concerned systems have certain characteristics, such as unknown nonlinear uncertainties, unknown dead-zones, unmodeled dynamics and without the direct measurements of state variables. In this paper, the neural networks (NNs) are employed to approximate the unknown nonlinear uncertainties, and then by representing the dead-zone as a time-varying system with a bounded disturbance. An NN state observer is designed to estimate the unmeasured states. Based on both backstepping design technique and a stochastic small-gain theorem, a robust adaptive NN output feedback control scheme is developed. It is proved that all the variables involved in the closed-loop system are input-state-practically stable in probability, and also have robustness to the unmodeled dynamics. Meanwhile, the observer errors and the output of the system can be regulated to a small neighborhood of the origin by selecting appropriate design parameters. Simulation examples are also provided to illustrate the effectiveness of the proposed approach.

Dynamic Output Feedback Control for Nonlinear Networked Control Systems with Random Packet Dropout and Random Delay

Directory of Open Access Journals (Sweden)

Shuiqing Yu

2013-01-01

Full Text Available This paper investigates the dynamic output feedback control for nonlinear networked control systems with both random packet dropout and random delay. Random packet dropout and random delay are modeled as two independent random variables. An observer-based dynamic output feedback controller is designed based upon the Lyapunov theory. The quantitative relationship of the dropout rate, transition probability matrix, and nonlinear level is derived by solving a set of linear matrix inequalities. Finally, an example is presented to illustrate the effectiveness of the proposed method.

Very Deep Convolutional Neural Networks for Morphologic Classification of Erythrocytes.

Science.gov (United States)

Durant, Thomas J S; Olson, Eben M; Schulz, Wade L; Torres, Richard

2017-12-01

Morphologic profiling of the erythrocyte population is a widely used and clinically valuable diagnostic modality, but one that relies on a slow manual process associated with significant labor cost and limited reproducibility. Automated profiling of erythrocytes from digital images by capable machine learning approaches would augment the throughput and value of morphologic analysis. To this end, we sought to evaluate the performance of leading implementation strategies for convolutional neural networks (CNNs) when applied to classification of erythrocytes based on morphology. Erythrocytes were manually classified into 1 of 10 classes using a custom-developed Web application. Using recent literature to guide architectural considerations for neural network design, we implemented a "very deep" CNN, consisting of >150 layers, with dense shortcut connections. The final database comprised 3737 labeled cells. Ensemble model predictions on unseen data demonstrated a harmonic mean of recall and precision metrics of 92.70% and 89.39%, respectively. Of the 748 cells in the test set, 23 misclassification errors were made, with a correct classification frequency of 90.60%, represented as a harmonic mean across the 10 morphologic classes. These findings indicate that erythrocyte morphology profiles could be measured with a high degree of accuracy with "very deep" CNNs. Further, these data support future efforts to expand classes and optimize practical performance in a clinical environment as a prelude to full implementation as a clinical tool. © 2017 American Association for Clinical Chemistry.

Classifying Radio Galaxies with the Convolutional Neural Network

Energy Technology Data Exchange (ETDEWEB)

Aniyan, A. K.; Thorat, K. [Department of Physics and Electronics, Rhodes University, Grahamstown (South Africa)

2017-06-01

We present the application of a deep machine learning technique to classify radio images of extended sources on a morphological basis using convolutional neural networks (CNN). In this study, we have taken the case of the Fanaroff–Riley (FR) class of radio galaxies as well as radio galaxies with bent-tailed morphology. We have used archival data from the Very Large Array (VLA)—Faint Images of the Radio Sky at Twenty Centimeters survey and existing visually classified samples available in the literature to train a neural network for morphological classification of these categories of radio sources. Our training sample size for each of these categories is ∼200 sources, which has been augmented by rotated versions of the same. Our study shows that CNNs can classify images of the FRI and FRII and bent-tailed radio galaxies with high accuracy (maximum precision at 95%) using well-defined samples and a “fusion classifier,” which combines the results of binary classifications, while allowing for a mechanism to find sources with unusual morphologies. The individual precision is highest for bent-tailed radio galaxies at 95% and is 91% and 75% for the FRI and FRII classes, respectively, whereas the recall is highest for FRI and FRIIs at 91% each, while the bent-tailed class has a recall of 79%. These results show that our results are comparable to that of manual classification, while being much faster. Finally, we discuss the computational and data-related challenges associated with the morphological classification of radio galaxies with CNNs.

Classifying Radio Galaxies with the Convolutional Neural Network

Science.gov (United States)

Aniyan, A. K.; Thorat, K.

2017-06-01

We present the application of a deep machine learning technique to classify radio images of extended sources on a morphological basis using convolutional neural networks (CNN). In this study, we have taken the case of the Fanaroff-Riley (FR) class of radio galaxies as well as radio galaxies with bent-tailed morphology. We have used archival data from the Very Large Array (VLA)—Faint Images of the Radio Sky at Twenty Centimeters survey and existing visually classified samples available in the literature to train a neural network for morphological classification of these categories of radio sources. Our training sample size for each of these categories is ˜200 sources, which has been augmented by rotated versions of the same. Our study shows that CNNs can classify images of the FRI and FRII and bent-tailed radio galaxies with high accuracy (maximum precision at 95%) using well-defined samples and a “fusion classifier,” which combines the results of binary classifications, while allowing for a mechanism to find sources with unusual morphologies. The individual precision is highest for bent-tailed radio galaxies at 95% and is 91% and 75% for the FRI and FRII classes, respectively, whereas the recall is highest for FRI and FRIIs at 91% each, while the bent-tailed class has a recall of 79%. These results show that our results are comparable to that of manual classification, while being much faster. Finally, we discuss the computational and data-related challenges associated with the morphological classification of radio galaxies with CNNs.

Classifying Radio Galaxies with the Convolutional Neural Network

International Nuclear Information System (INIS)

Aniyan, A. K.; Thorat, K.

2017-01-01

We present the application of a deep machine learning technique to classify radio images of extended sources on a morphological basis using convolutional neural networks (CNN). In this study, we have taken the case of the Fanaroff–Riley (FR) class of radio galaxies as well as radio galaxies with bent-tailed morphology. We have used archival data from the Very Large Array (VLA)—Faint Images of the Radio Sky at Twenty Centimeters survey and existing visually classified samples available in the literature to train a neural network for morphological classification of these categories of radio sources. Our training sample size for each of these categories is ∼200 sources, which has been augmented by rotated versions of the same. Our study shows that CNNs can classify images of the FRI and FRII and bent-tailed radio galaxies with high accuracy (maximum precision at 95%) using well-defined samples and a “fusion classifier,” which combines the results of binary classifications, while allowing for a mechanism to find sources with unusual morphologies. The individual precision is highest for bent-tailed radio galaxies at 95% and is 91% and 75% for the FRI and FRII classes, respectively, whereas the recall is highest for FRI and FRIIs at 91% each, while the bent-tailed class has a recall of 79%. These results show that our results are comparable to that of manual classification, while being much faster. Finally, we discuss the computational and data-related challenges associated with the morphological classification of radio galaxies with CNNs.

A universal concept based on cellular neural networks for ultrafast and flexible solving of differential equations.

Science.gov (United States)

Chedjou, Jean Chamberlain; Kyamakya, Kyandoghere

2015-04-01

This paper develops and validates a comprehensive and universally applicable computational concept for solving nonlinear differential equations (NDEs) through a neurocomputing concept based on cellular neural networks (CNNs). High-precision, stability, convergence, and lowest-possible memory requirements are ensured by the CNN processor architecture. A significant challenge solved in this paper is that all these cited computing features are ensured in all system-states (regular or chaotic ones) and in all bifurcation conditions that may be experienced by NDEs.One particular quintessence of this paper is to develop and demonstrate a solver concept that shows and ensures that CNN processors (realized either in hardware or in software) are universal solvers of NDE models. The solving logic or algorithm of given NDEs (possible examples are: Duffing, Mathieu, Van der Pol, Jerk, Chua, Rössler, Lorenz, Burgers, and the transport equations) through a CNN processor system is provided by a set of templates that are computed by our comprehensive templates calculation technique that we call nonlinear adaptive optimization. This paper is therefore a significant contribution and represents a cutting-edge real-time computational engineering approach, especially while considering the various scientific and engineering applications of this ultrafast, energy-and-memory-efficient, and high-precise NDE solver concept. For illustration purposes, three NDE models are demonstratively solved, and related CNN templates are derived and used: the periodically excited Duffing equation, the Mathieu equation, and the transport equation.

Quaternion-based adaptive output feedback attitude control of spacecraft using Chebyshev neural networks.

Science.gov (United States)

Zou, An-Min; Dev Kumar, Krishna; Hou, Zeng-Guang

2010-09-01

This paper investigates the problem of output feedback attitude control of an uncertain spacecraft. Two robust adaptive output feedback controllers based on Chebyshev neural networks (CNN) termed adaptive neural networks (NN) controller-I and adaptive NN controller-II are proposed for the attitude tracking control of spacecraft. The four-parameter representations (quaternion) are employed to describe the spacecraft attitude for global representation without singularities. The nonlinear reduced-order observer is used to estimate the derivative of the spacecraft output, and the CNN is introduced to further improve the control performance through approximating the spacecraft attitude motion. The implementation of the basis functions of the CNN used in the proposed controllers depends only on the desired signals, and the smooth robust compensator using the hyperbolic tangent function is employed to counteract the CNN approximation errors and external disturbances. The adaptive NN controller-II can efficiently avoid the over-estimation problem (i.e., the bound of the CNNs output is much larger than that of the approximated unknown function, and hence, the control input may be very large) existing in the adaptive NN controller-I. Both adaptive output feedback controllers using CNN can guarantee that all signals in the resulting closed-loop system are uniformly ultimately bounded. For performance comparisons, the standard adaptive controller using the linear parameterization of spacecraft attitude motion is also developed. Simulation studies are presented to show the advantages of the proposed CNN-based output feedback approach over the standard adaptive output feedback approach.

Nonlinear dynamics and complexity

CERN Document Server

Luo, Albert; Fu, Xilin

2014-01-01

This important collection presents recent advances in nonlinear dynamics including analytical solutions, chaos in Hamiltonian systems, time-delay, uncertainty, and bio-network dynamics. Nonlinear Dynamics and Complexity equips readers to appreciate this increasingly main-stream approach to understanding complex phenomena in nonlinear systems as they are examined in a broad array of disciplines. The book facilitates a better understanding of the mechanisms and phenomena in nonlinear dynamics and develops the corresponding mathematical theory to apply nonlinear design to practical engineering.

Adaptive Neural Networks Decentralized FTC Design for Nonstrict-Feedback Nonlinear Interconnected Large-Scale Systems Against Actuator Faults.

Science.gov (United States)

Li, Yongming; Tong, Shaocheng

The problem of active fault-tolerant control (FTC) is investigated for the large-scale nonlinear systems in nonstrict-feedback form. The nonstrict-feedback nonlinear systems considered in this paper consist of unstructured uncertainties, unmeasured states, unknown interconnected terms, and actuator faults (e.g., bias fault and gain fault). A state observer is designed to solve the unmeasurable state problem. Neural networks (NNs) are used to identify the unknown lumped nonlinear functions so that the problems of unstructured uncertainties and unknown interconnected terms can be solved. By combining the adaptive backstepping design principle with the combination Nussbaum gain function property, a novel NN adaptive output-feedback FTC approach is developed. The proposed FTC controller can guarantee that all signals in all subsystems are bounded, and the tracking errors for each subsystem converge to a small neighborhood of zero. Finally, numerical results of practical examples are presented to further demonstrate the effectiveness of the proposed control strategy.The problem of active fault-tolerant control (FTC) is investigated for the large-scale nonlinear systems in nonstrict-feedback form. The nonstrict-feedback nonlinear systems considered in this paper consist of unstructured uncertainties, unmeasured states, unknown interconnected terms, and actuator faults (e.g., bias fault and gain fault). A state observer is designed to solve the unmeasurable state problem. Neural networks (NNs) are used to identify the unknown lumped nonlinear functions so that the problems of unstructured uncertainties and unknown interconnected terms can be solved. By combining the adaptive backstepping design principle with the combination Nussbaum gain function property, a novel NN adaptive output-feedback FTC approach is developed. The proposed FTC controller can guarantee that all signals in all subsystems are bounded, and the tracking errors for each subsystem converge to a small

Convolutional Neural Networks for Text Categorization: Shallow Word-level vs. Deep Character-level

OpenAIRE

Johnson, Rie; Zhang, Tong

2016-01-01

This paper reports the performances of shallow word-level convolutional neural networks (CNN), our earlier work (2015), on the eight datasets with relatively large training data that were used for testing the very deep character-level CNN in Conneau et al. (2016). Our findings are as follows. The shallow word-level CNNs achieve better error rates than the error rates reported in Conneau et al., though the results should be interpreted with some consideration due to the unique pre-processing o...

Heart Rate Estimated from Body Movements at Six Degrees of Freedom by Convolutional Neural Networks.

Science.gov (United States)

Lee, Hyunwoo; Whang, Mincheol

2018-05-01

Cardiac activity has been monitored continuously in daily life by virtue of advanced medical instruments with microelectromechanical system (MEMS) technology. Seismocardiography (SCG) has been considered to be free from the burden of measurement for cardiac activity, but it has been limited in its application in daily life. The most important issues regarding SCG are to overcome the limitations of motion artifacts due to the sensitivity of motion sensor. Although novel adaptive filters for noise cancellation have been developed, they depend on the researcher’s subjective decision. Convolutional neural networks (CNNs) can extract significant features from data automatically without a researcher’s subjective decision, so that signal processing has been recently replaced as CNNs. Thus, this study aimed to develop a novel method to enhance heart rate estimation from thoracic movement by CNNs. Thoracic movement was measured by six-axis accelerometer and gyroscope signals using a wearable sensor that can be worn by simply clipping on clothes. The dataset was collected from 30 participants (15 males, 15 females) using 12 measurement conditions according to two physical conditions (i.e., relaxed and aroused conditions), three body postures (i.e., sitting, standing, and supine), and six movement speeds (i.e., 3.2, 4.5, 5.8, 6.4, 8.5, and 10.3 km/h). The motion data (i.e., six-axis accelerometer and gyroscope) and heart rate (i.e., electrocardiogram (ECG)) were determined as the input data and labels in the dataset, respectively. The CNN model was developed based on VGG Net and optimized by testing according to network depth and data augmentation. The ensemble network of the VGG-16 without data augmentation and the VGG-19 with data augmentation was determined as optimal architecture for generalization. As a result, the proposed method showed higher accuracy than the previous SCG method using signal processing in most measurement conditions. The three main contributions

Heart Rate Estimated from Body Movements at Six Degrees of Freedom by Convolutional Neural Networks

Directory of Open Access Journals (Sweden)

Hyunwoo Lee

2018-05-01

Full Text Available Cardiac activity has been monitored continuously in daily life by virtue of advanced medical instruments with microelectromechanical system (MEMS technology. Seismocardiography (SCG has been considered to be free from the burden of measurement for cardiac activity, but it has been limited in its application in daily life. The most important issues regarding SCG are to overcome the limitations of motion artifacts due to the sensitivity of motion sensor. Although novel adaptive filters for noise cancellation have been developed, they depend on the researcher’s subjective decision. Convolutional neural networks (CNNs can extract significant features from data automatically without a researcher’s subjective decision, so that signal processing has been recently replaced as CNNs. Thus, this study aimed to develop a novel method to enhance heart rate estimation from thoracic movement by CNNs. Thoracic movement was measured by six-axis accelerometer and gyroscope signals using a wearable sensor that can be worn by simply clipping on clothes. The dataset was collected from 30 participants (15 males, 15 females using 12 measurement conditions according to two physical conditions (i.e., relaxed and aroused conditions, three body postures (i.e., sitting, standing, and supine, and six movement speeds (i.e., 3.2, 4.5, 5.8, 6.4, 8.5, and 10.3 km/h. The motion data (i.e., six-axis accelerometer and gyroscope and heart rate (i.e., electrocardiogram (ECG were determined as the input data and labels in the dataset, respectively. The CNN model was developed based on VGG Net and optimized by testing according to network depth and data augmentation. The ensemble network of the VGG-16 without data augmentation and the VGG-19 with data augmentation was determined as optimal architecture for generalization. As a result, the proposed method showed higher accuracy than the previous SCG method using signal processing in most measurement conditions. The three main

Expectation propagation for large scale Bayesian inference of non-linear molecular networks from perturbation data.

Science.gov (United States)

Narimani, Zahra; Beigy, Hamid; Ahmad, Ashar; Masoudi-Nejad, Ali; Fröhlich, Holger

2017-01-01

Inferring the structure of molecular networks from time series protein or gene expression data provides valuable information about the complex biological processes of the cell. Causal network structure inference has been approached using different methods in the past. Most causal network inference techniques, such as Dynamic Bayesian Networks and ordinary differential equations, are limited by their computational complexity and thus make large scale inference infeasible. This is specifically true if a Bayesian framework is applied in order to deal with the unavoidable uncertainty about the correct model. We devise a novel Bayesian network reverse engineering approach using ordinary differential equations with the ability to include non-linearity. Besides modeling arbitrary, possibly combinatorial and time dependent perturbations with unknown targets, one of our main contributions is the use of Expectation Propagation, an algorithm for approximate Bayesian inference over large scale network structures in short computation time. We further explore the possibility of integrating prior knowledge into network inference. We evaluate the proposed model on DREAM4 and DREAM8 data and find it competitive against several state-of-the-art existing network inference methods.

Finite-time generalized function matrix projective lag synchronization of coupled dynamical networks with different dimensions via the double power function nonlinear feedback control method

International Nuclear Information System (INIS)

Dai, Hao; Si, Gangquan; Jia, Lixin; Zhang, Yanbin

2014-01-01

This paper investigates the problem of finite-time generalized function matrix projective lag synchronization between two different coupled dynamical networks with different dimensions of network nodes. The double power function nonlinear feedback control method is proposed in this paper to guarantee that the state trajectories of the response network converge to the state trajectories of the drive network according to a function matrix in a given finite time. Furthermore, in comparison with the traditional nonlinear feedback control method, the new method improves the synchronization efficiency, and shortens the finite synchronization time. Numerical simulation results are presented to illustrate the effectiveness of this method. (papers)

Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection

Science.gov (United States)

Cabrera-Vives, Guillermo; Reyes, Ignacio; Förster, Francisco; Estévez, Pablo A.; Maureira, Juan-Carlos

2017-02-01

We introduce Deep-HiTS, a rotation-invariant convolutional neural network (CNN) model for classifying images of transient candidates into artifacts or real sources for the High cadence Transient Survey (HiTS). CNNs have the advantage of learning the features automatically from the data while achieving high performance. We compare our CNN model against a feature engineering approach using random forests (RFs). We show that our CNN significantly outperforms the RF model, reducing the error by almost half. Furthermore, for a fixed number of approximately 2000 allowed false transient candidates per night, we are able to reduce the misclassified real transients by approximately one-fifth. To the best of our knowledge, this is the first time CNNs have been used to detect astronomical transient events. Our approach will be very useful when processing images from next generation instruments such as the Large Synoptic Survey Telescope. We have made all our code and data available to the community for the sake of allowing further developments and comparisons at https://github.com/guille-c/Deep-HiTS. Deep-HiTS is licensed under the terms of the GNU General Public License v3.0.

Unsupervised Document Embedding With CNNs

OpenAIRE

Liu, Chundi; Zhao, Shunan; Volkovs, Maksims

2017-01-01

We propose a new model for unsupervised document embedding. Leading existing approaches either require complex inference or use recurrent neural networks (RNN) that are difficult to parallelize. We take a different route and develop a convolutional neural network (CNN) embedding model. Our CNN architecture is fully parallelizable resulting in over 10x speedup in inference time over RNN models. Parallelizable architecture enables to train deeper models where each successive layer has increasin...

Convolutional neural networks for segmentation and object detection of human semen

DEFF Research Database (Denmark)

Nissen, Malte Stær; Krause, Oswin; Almstrup, Kristian

2017-01-01

to clutter. Our results indicate that training on full images is superior to training on patches when class-skew is properly handled. Full image training including up-sampling during training proves to be beneficial in deep CNNs for pixel wise accuracy and detection performance. Predicted sperm cells...... are found by using connected components on the CNN predictions. We investigate optimization of a threshold parameter on the size of detected components. Our best network achieves 93.87% precision and 91.89% recall on our test dataset after thresholding outperforming a classical image analysis approach....

SOS based robust H(∞) fuzzy dynamic output feedback control of nonlinear networked control systems.

Science.gov (United States)

Chae, Seunghwan; Nguang, Sing Kiong

2014-07-01

In this paper, a methodology for designing a fuzzy dynamic output feedback controller for discrete-time nonlinear networked control systems is presented where the nonlinear plant is modelled by a Takagi-Sugeno fuzzy model and the network-induced delays by a finite state Markov process. The transition probability matrix for the Markov process is allowed to be partially known, providing a more practical consideration of the real world. Furthermore, the fuzzy controller's membership functions and premise variables are not assumed to be the same as the plant's membership functions and premise variables, that is, the proposed approach can handle the case, when the premise of the plant are not measurable or delayed. The membership functions of the plant and the controller are approximated as polynomial functions, then incorporated into the controller design. Sufficient conditions for the existence of the controller are derived in terms of sum of square inequalities, which are then solved by YALMIP. Finally, a numerical example is used to demonstrate the validity of the proposed methodology.

Prediction of municipal solid waste generation using nonlinear autoregressive network.

Science.gov (United States)

Younes, Mohammad K; Nopiah, Z M; Basri, N E Ahmad; Basri, H; Abushammala, Mohammed F M; Maulud, K N A

2015-12-01

Most of the developing countries have solid waste management problems. Solid waste strategic planning requires accurate prediction of the quality and quantity of the generated waste. In developing countries, such as Malaysia, the solid waste generation rate is increasing rapidly, due to population growth and new consumption trends that characterize society. This paper proposes an artificial neural network (ANN) approach using feedforward nonlinear autoregressive network with exogenous inputs (NARX) to predict annual solid waste generation in relation to demographic and economic variables like population number, gross domestic product, electricity demand per capita and employment and unemployment numbers. In addition, variable selection procedures are also developed to select a significant explanatory variable. The model evaluation was performed using coefficient of determination (R(2)) and mean square error (MSE). The optimum model that produced the lowest testing MSE (2.46) and the highest R(2) (0.97) had three inputs (gross domestic product, population and employment), eight neurons and one lag in the hidden layer, and used Fletcher-Powell's conjugate gradient as the training algorithm.

Decoupled ARX and RBF Neural Network Modeling Using PCA and GA Optimization for Nonlinear Distributed Parameter Systems.

Science.gov (United States)

Zhang, Ridong; Tao, Jili; Lu, Renquan; Jin, Qibing

2018-02-01

Modeling of distributed parameter systems is difficult because of their nonlinearity and infinite-dimensional characteristics. Based on principal component analysis (PCA), a hybrid modeling strategy that consists of a decoupled linear autoregressive exogenous (ARX) model and a nonlinear radial basis function (RBF) neural network model are proposed. The spatial-temporal output is first divided into a few dominant spatial basis functions and finite-dimensional temporal series by PCA. Then, a decoupled ARX model is designed to model the linear dynamics of the dominant modes of the time series. The nonlinear residual part is subsequently parameterized by RBFs, where genetic algorithm is utilized to optimize their hidden layer structure and the parameters. Finally, the nonlinear spatial-temporal dynamic system is obtained after the time/space reconstruction. Simulation results of a catalytic rod and a heat conduction equation demonstrate the effectiveness of the proposed strategy compared to several other methods.

An extended Kalman filtering approach to modeling nonlinear dynamic gene regulatory networks via short gene expression time series.

Science.gov (United States)

Wang, Zidong; Liu, Xiaohui; Liu, Yurong; Liang, Jinling; Vinciotti, Veronica

2009-01-01

In this paper, the extended Kalman filter (EKF) algorithm is applied to model the gene regulatory network from gene time series data. The gene regulatory network is considered as a nonlinear dynamic stochastic model that consists of the gene measurement equation and the gene regulation equation. After specifying the model structure, we apply the EKF algorithm for identifying both the model parameters and the actual value of gene expression levels. It is shown that the EKF algorithm is an online estimation algorithm that can identify a large number of parameters (including parameters of nonlinear functions) through iterative procedure by using a small number of observations. Four real-world gene expression data sets are employed to demonstrate the effectiveness of the EKF algorithm, and the obtained models are evaluated from the viewpoint of bioinformatics.

Nonlinearly Activated Neural Network for Solving Time-Varying Complex Sylvester Equation.

Science.gov (United States)

Li, Shuai; Li, Yangming

2013-10-28

The Sylvester equation is often encountered in mathematics and control theory. For the general time-invariant Sylvester equation problem, which is defined in the domain of complex numbers, the Bartels-Stewart algorithm and its extensions are effective and widely used with an O(n³) time complexity. When applied to solving the time-varying Sylvester equation, the computation burden increases intensively with the decrease of sampling period and cannot satisfy continuous realtime calculation requirements. For the special case of the general Sylvester equation problem defined in the domain of real numbers, gradient-based recurrent neural networks are able to solve the time-varying Sylvester equation in real time, but there always exists an estimation error while a recently proposed recurrent neural network by Zhang et al [this type of neural network is called Zhang neural network (ZNN)] converges to the solution ideally. The advancements in complex-valued neural networks cast light to extend the existing real-valued ZNN for solving the time-varying real-valued Sylvester equation to its counterpart in the domain of complex numbers. In this paper, a complex-valued ZNN for solving the complex-valued Sylvester equation problem is investigated and the global convergence of the neural network is proven with the proposed nonlinear complex-valued activation functions. Moreover, a special type of activation function with a core function, called sign-bi-power function, is proven to enable the ZNN to converge in finite time, which further enhances its advantage in online processing. In this case, the upper bound of the convergence time is also derived analytically. Simulations are performed to evaluate and compare the performance of the neural network with different parameters and activation functions. Both theoretical analysis and numerical simulations validate the effectiveness of the proposed method.

Distributed synchronization of networked drive-response systems: A nonlinear fixed-time protocol.

Science.gov (United States)

Zhao, Wen; Liu, Gang; Ma, Xi; He, Bing; Dong, Yunfeng

2017-11-01

The distributed synchronization of networked drive-response systems is investigated in this paper. A novel nonlinear protocol is proposed to ensure that the tracking errors converge to zeros in a fixed-time. By comparison with previous synchronization methods, the present method considers more practical conditions and the synchronization time is not dependent of arbitrary initial conditions but can be offline pre-assign according to the task assignment. Finally, the feasibility and validity of the presented protocol have been illustrated by a numerical simulation. Copyright © 2017. Published by Elsevier Ltd.

Comparison of machine learned approaches for thyroid nodule characterization from shear wave elastography images

Science.gov (United States)

Pereira, Carina; Dighe, Manjiri; Alessio, Adam M.

2018-02-01

Various Computer Aided Diagnosis (CAD) systems have been developed that characterize thyroid nodules using the features extracted from the B-mode ultrasound images and Shear Wave Elastography images (SWE). These features, however, are not perfect predictors of malignancy. In other domains, deep learning techniques such as Convolutional Neural Networks (CNNs) have outperformed conventional feature extraction based machine learning approaches. In general, fully trained CNNs require substantial volumes of data, motivating several efforts to use transfer learning with pre-trained CNNs. In this context, we sought to compare the performance of conventional feature extraction, fully trained CNNs, and transfer learning based, pre-trained CNNs for the detection of thyroid malignancy from ultrasound images. We compared these approaches applied to a data set of 964 B-mode and SWE images from 165 patients. The data were divided into 80% training/validation and 20% testing data. The highest accuracies achieved on the testing data for the conventional feature extraction, fully trained CNN, and pre-trained CNN were 0.80, 0.75, and 0.83 respectively. In this application, classification using a pre-trained network yielded the best performance, potentially due to the relatively limited sample size and sub-optimal architecture for the fully trained CNN.

Evaluation of fatigue cracks using nonlinearities of acousto-ultrasonic waves acquired by an active sensor network

International Nuclear Information System (INIS)

Zhou, Chao; Hong, Ming; Su, Zhongqing; Wang, Qiang; Cheng, Li

2013-01-01

There has been increasing interest in using the nonlinear features of acousto-ultrasonic (AU) waves to detect damage onset (e.g., micro-fatigue cracks) due to their high sensitivity to damage with small dimensions. However, most existing approaches are able to infer the existence of fatigue damage qualitatively, but fail to further ascertain its location and severity. A damage characterization approach, in conjunction with the use of an active piezoelectric sensor network, was established, capable of evaluating fatigue cracks in a quantitative manner (including the co-presence of multiple fatigue cracks, and their individual locations and severities). Fundamental investigations, using both experiment and enhanced finite element analysis dedicated to the simulation of nonlinear AU waves, were carried out to link the accumulation of nonlinearities extracted from high-order AU waves to the characteristic parameters of a fatigue crack. A probability-based diagnostic imaging algorithm was developed, facilitating an intuitive presentation of identification results in images. The approach was verified experimentally by evaluating multi-fatigue cracks near rivet holes of a fatigued aluminum plate, showing satisfactory precision in characterizing real, barely visible fatigue cracks. Compared with existing methods, this approach innovatively (i) uses permanently integrated active sensor networks, conducive to automatic and online health monitoring; (ii) characterizes fatigue cracks at a quantitative level; (iii) allows detection of multiple fatigue cracks; and (iv) visualizes identification results in intuitive images. (paper)

Nonlinear Equalization in 40/112/224 Gbit/s Mixed Line Rate 15-Channel DP-QPSK and DP-16QAM Contiguous Spectrum Based Networks

DEFF Research Database (Denmark)

Asif, Rameez

2014-01-01

We evaluated that in-line non-linear compensation schemes decrease the com- plexity of digital back-propagation and enhance the perfor mance of 40/112/224Gbit/s mixed line rate network. Both grouped and un-grouped spectral all ocation schemes are investigated.......We evaluated that in-line non-linear compensation schemes decrease the com- plexity of digital back-propagation and enhance the perfor mance of 40/112/224Gbit/s mixed line rate network. Both grouped and un-grouped spectral all ocation schemes are investigated....

Nonlinear Dynamic Surface Control of Chaos in Permanent Magnet Synchronous Motor Based on the Minimum Weights of RBF Neural Network

Directory of Open Access Journals (Sweden)

Shaohua Luo

2014-01-01

Full Text Available This paper is concerned with the problem of the nonlinear dynamic surface control (DSC of chaos based on the minimum weights of RBF neural network for the permanent magnet synchronous motor system (PMSM wherein the unknown parameters, disturbances, and chaos are presented. RBF neural network is used to approximate the nonlinearities and an adaptive law is employed to estimate unknown parameters. Then, a simple and effective controller is designed by introducing dynamic surface control technique on the basis of first-order filters. Asymptotically tracking stability in the sense of uniformly ultimate boundedness is achieved in a short time. Finally, the performance of the proposed controller is testified through simulation results.

Latest developments in image processing for the next generation of devices with a view on DEMO

International Nuclear Information System (INIS)

Murari, A.; Vega, J.; Mazon, D.; Arena, P.; Craciunescu, T.; Gabellieri, L.; Gelfusa, M.; Pacella, D.; Palazzo, S.; Romano, A.

2012-01-01

Highlights: ► Pattern recognition methods have been successfully applied to retrieve frames in massime databases. ► The technology of Cellular Nonlinear Networks (CNNs) has been upgraded to solve space variant problem. ► The CNNs have then been successfully applied to various tasks, from the real time hot spot detection to the automatic identification of instabilities. ► The method of the optical flow permits to derive information about the speed of the objects moving in the frames of a single camera. ► Since the next generation of devices will emit a lot of SXR radiation, also from the edge, new technologies (Gas Electron Multiplier detectors and policapillary lenses) are being developed to perform imaging over this region of the spectrum for a global view of the entire plasma column. - Abstract: In magnetic confinement fusion devices the use of cameras, both visible and infrared, has increased very significantly in the last years. The large amount of data (in the range of tens of Gbytes per shot) and the difficulty of the analysis tasks (ambiguity, ill posed problems, etc.), require new solutions. The technology of Cellular Nonlinear Networks (CNNs) has been successfully applied to various tasks, from the real time hot spot detection to the automatic identification of instabilities. The accuracy obtained is comparable to the one of more traditional serial algorithms but the CCNs guarantee deterministic computational times independently from the image contents. Moreover the latest developments have allowed obtaining these results also in the case of space variant image analysis, without compromising the computational speed (of the order of ten thousand frames per second). The method of the optical flow permits to derive information about the speed of the objects moving in the frames of a single camera. The results of previous applications have been so successful that the approach has been extended to videos in compressed format (MPEG) to reduce the

Binary patterns encoded convolutional neural networks for texture recognition and remote sensing scene classification

Science.gov (United States)

Anwer, Rao Muhammad; Khan, Fahad Shahbaz; van de Weijer, Joost; Molinier, Matthieu; Laaksonen, Jorma

2018-04-01

Designing discriminative powerful texture features robust to realistic imaging conditions is a challenging computer vision problem with many applications, including material recognition and analysis of satellite or aerial imagery. In the past, most texture description approaches were based on dense orderless statistical distribution of local features. However, most recent approaches to texture recognition and remote sensing scene classification are based on Convolutional Neural Networks (CNNs). The de facto practice when learning these CNN models is to use RGB patches as input with training performed on large amounts of labeled data (ImageNet). In this paper, we show that Local Binary Patterns (LBP) encoded CNN models, codenamed TEX-Nets, trained using mapped coded images with explicit LBP based texture information provide complementary information to the standard RGB deep models. Additionally, two deep architectures, namely early and late fusion, are investigated to combine the texture and color information. To the best of our knowledge, we are the first to investigate Binary Patterns encoded CNNs and different deep network fusion architectures for texture recognition and remote sensing scene classification. We perform comprehensive experiments on four texture recognition datasets and four remote sensing scene classification benchmarks: UC-Merced with 21 scene categories, WHU-RS19 with 19 scene classes, RSSCN7 with 7 categories and the recently introduced large scale aerial image dataset (AID) with 30 aerial scene types. We demonstrate that TEX-Nets provide complementary information to standard RGB deep model of the same network architecture. Our late fusion TEX-Net architecture always improves the overall performance compared to the standard RGB network on both recognition problems. Furthermore, our final combination leads to consistent improvement over the state-of-the-art for remote sensing scene classification.

Robust FDI for a Class of Nonlinear Networked Systems with ROQs

Directory of Open Access Journals (Sweden)

An-quan Sun

2014-01-01

Full Text Available This paper considers the robust fault detection and isolation (FDI problem for a class of nonlinear networked systems (NSs with randomly occurring quantisations (ROQs. After vector augmentation, Lyapunov function is introduced to ensure the asymptotically mean-square stability of fault detection system. By transforming the quantisation effects into sector-bounded parameter uncertainties, sufficient conditions ensuring the existence of fault detection filter are proposed, which can reduce the difference between output residuals and fault signals as small as possible under H∞ framework. Finally, an example linearized from a vehicle system is introduced to show the efficiency of the proposed fault detection filter.

OKVAR-Boost: a novel boosting algorithm to infer nonlinear dynamics and interactions in gene regulatory networks.

Science.gov (United States)

Lim, Néhémy; Senbabaoglu, Yasin; Michailidis, George; d'Alché-Buc, Florence

2013-06-01

Reverse engineering of gene regulatory networks remains a central challenge in computational systems biology, despite recent advances facilitated by benchmark in silico challenges that have aided in calibrating their performance. A number of approaches using either perturbation (knock-out) or wild-type time-series data have appeared in the literature addressing this problem, with the latter using linear temporal models. Nonlinear dynamical models are particularly appropriate for this inference task, given the generation mechanism of the time-series data. In this study, we introduce a novel nonlinear autoregressive model based on operator-valued kernels that simultaneously learns the model parameters, as well as the network structure. A flexible boosting algorithm (OKVAR-Boost) that shares features from L2-boosting and randomization-based algorithms is developed to perform the tasks of parameter learning and network inference for the proposed model. Specifically, at each boosting iteration, a regularized Operator-valued Kernel-based Vector AutoRegressive model (OKVAR) is trained on a random subnetwork. The final model consists of an ensemble of such models. The empirical estimation of the ensemble model's Jacobian matrix provides an estimation of the network structure. The performance of the proposed algorithm is first evaluated on a number of benchmark datasets from the DREAM3 challenge and then on real datasets related to the In vivo Reverse-Engineering and Modeling Assessment (IRMA) and T-cell networks. The high-quality results obtained strongly indicate that it outperforms existing approaches. The OKVAR-Boost Matlab code is available as the archive: http://amis-group.fr/sourcecode-okvar-boost/OKVARBoost-v1.0.zip. Supplementary data are available at Bioinformatics online.

Non-intrusive reduced order modeling of nonlinear problems using neural networks

Science.gov (United States)

Hesthaven, J. S.; Ubbiali, S.

2018-06-01

We develop a non-intrusive reduced basis (RB) method for parametrized steady-state partial differential equations (PDEs). The method extracts a reduced basis from a collection of high-fidelity solutions via a proper orthogonal decomposition (POD) and employs artificial neural networks (ANNs), particularly multi-layer perceptrons (MLPs), to accurately approximate the coefficients of the reduced model. The search for the optimal number of neurons and the minimum amount of training samples to avoid overfitting is carried out in the offline phase through an automatic routine, relying upon a joint use of the Latin hypercube sampling (LHS) and the Levenberg-Marquardt (LM) training algorithm. This guarantees a complete offline-online decoupling, leading to an efficient RB method - referred to as POD-NN - suitable also for general nonlinear problems with a non-affine parametric dependence. Numerical studies are presented for the nonlinear Poisson equation and for driven cavity viscous flows, modeled through the steady incompressible Navier-Stokes equations. Both physical and geometrical parametrizations are considered. Several results confirm the accuracy of the POD-NN method and show the substantial speed-up enabled at the online stage as compared to a traditional RB strategy.

Zero-sum two-player game theoretic formulation of affine nonlinear discrete-time systems using neural networks.

Science.gov (United States)

Mehraeen, Shahab; Dierks, Travis; Jagannathan, S; Crow, Mariesa L

2013-12-01

In this paper, the nearly optimal solution for discrete-time (DT) affine nonlinear control systems in the presence of partially unknown internal system dynamics and disturbances is considered. The approach is based on successive approximate solution of the Hamilton-Jacobi-Isaacs (HJI) equation, which appears in optimal control. Successive approximation approach for updating control and disturbance inputs for DT nonlinear affine systems are proposed. Moreover, sufficient conditions for the convergence of the approximate HJI solution to the saddle point are derived, and an iterative approach to approximate the HJI equation using a neural network (NN) is presented. Then, the requirement of full knowledge of the internal dynamics of the nonlinear DT system is relaxed by using a second NN online approximator. The result is a closed-loop optimal NN controller via offline learning. A numerical example is provided illustrating the effectiveness of the approach.

Identifiability of large-scale non-linear dynamic network models applied to the ADM1-case study.

Science.gov (United States)

Nimmegeers, Philippe; Lauwers, Joost; Telen, Dries; Logist, Filip; Impe, Jan Van

2017-06-01

In this work, both the structural and practical identifiability of the Anaerobic Digestion Model no. 1 (ADM1) is investigated, which serves as a relevant case study of large non-linear dynamic network models. The structural identifiability is investigated using the probabilistic algorithm, adapted to deal with the specifics of the case study (i.e., a large-scale non-linear dynamic system of differential and algebraic equations). The practical identifiability is analyzed using a Monte Carlo parameter estimation procedure for a 'non-informative' and 'informative' experiment, which are heuristically designed. The model structure of ADM1 has been modified by replacing parameters by parameter combinations, to provide a generally locally structurally identifiable version of ADM1. This means that in an idealized theoretical situation, the parameters can be estimated accurately. Furthermore, the generally positive structural identifiability results can be explained from the large number of interconnections between the states in the network structure. This interconnectivity, however, is also observed in the parameter estimates, making uncorrelated parameter estimations in practice difficult. Copyright © 2017. Published by Elsevier Inc.

On the sample complexity of learning for networks of spiking neurons with nonlinear synaptic interactions.

Science.gov (United States)

Schmitt, Michael

2004-09-01

We study networks of spiking neurons that use the timing of pulses to encode information. Nonlinear interactions model the spatial groupings of synapses on the neural dendrites and describe the computations performed at local branches. Within a theoretical framework of learning we analyze the question of how many training examples these networks must receive to be able to generalize well. Bounds for this sample complexity of learning can be obtained in terms of a combinatorial parameter known as the pseudodimension. This dimension characterizes the computational richness of a neural network and is given in terms of the number of network parameters. Two types of feedforward architectures are considered: constant-depth networks and networks of unconstrained depth. We derive asymptotically tight bounds for each of these network types. Constant depth networks are shown to have an almost linear pseudodimension, whereas the pseudodimension of general networks is quadratic. Networks of spiking neurons that use temporal coding are becoming increasingly more important in practical tasks such as computer vision, speech recognition, and motor control. The question of how well these networks generalize from a given set of training examples is a central issue for their successful application as adaptive systems. The results show that, although coding and computation in these networks is quite different and in many cases more powerful, their generalization capabilities are at least as good as those of traditional neural network models.

Neural network-based robust actuator fault diagnosis for a non-linear multi-tank system.

Science.gov (United States)

Mrugalski, Marcin; Luzar, Marcel; Pazera, Marcin; Witczak, Marcin; Aubrun, Christophe

2016-03-01

The paper is devoted to the problem of the robust actuator fault diagnosis of the dynamic non-linear systems. In the proposed method, it is assumed that the diagnosed system can be modelled by the recurrent neural network, which can be transformed into the linear parameter varying form. Such a system description allows developing the designing scheme of the robust unknown input observer within H∞ framework for a class of non-linear systems. The proposed approach is designed in such a way that a prescribed disturbance attenuation level is achieved with respect to the actuator fault estimation error, while guaranteeing the convergence of the observer. The application of the robust unknown input observer enables actuator fault estimation, which allows applying the developed approach to the fault tolerant control tasks. Copyright © 2016 ISA. Published by Elsevier Ltd. All rights reserved.

Exponential synchronization of two nonlinearly non-delayed and delayed coupled complex dynamical networks

International Nuclear Information System (INIS)

Zheng Song

2012-01-01

In this paper, the exponential synchronization between two nonlinearly coupled complex networks with non-delayed and delayed coupling is investigated with Lyapunov-Krasovskii-type functionals. Based on the stability analysis of the impulsive differential equation and the linear matrix inequality, sufficient delay-dependent conditions for exponential synchronization are derived, and a linear impulsive controller and simple updated laws are also designed. Furthermore, the coupling matrices need not be symmetric or irreducible. Numerical examples are presented to verify the effectiveness and correctness of the synchronization criteria obtained.

Global exponential stability of BAM neural networks with transmission delays and nonlinear impulses

International Nuclear Information System (INIS)

Huang Zhenkun; Xia Yonghui

2008-01-01

In this paper, a class of bidirectional associative memory (BAM) networks with transmission delays and nonlinear impulses are studied. Some new sufficient conditions are established for the existence and global exponential stability of a unique equilibrium, which generalize and improve the previously known results. The sufficient conditions are easy to verify and when the impulsive jumps are linear or absent the results reduce to those of common impulsive or non-impulsive systems. Finally, an example is given to show the feasibility and effectiveness of our results

Generalized Nonlinear Yule Models

OpenAIRE

Lansky, Petr; Polito, Federico; Sacerdote, Laura

2016-01-01

With the aim of considering models with persistent memory we propose a fractional nonlinear modification of the classical Yule model often studied in the context of macrovolution. Here the model is analyzed and interpreted in the framework of the development of networks such as the World Wide Web. Nonlinearity is introduced by replacing the linear birth process governing the growth of the in-links of each specific webpage with a fractional nonlinear birth process with completely general birth...

Constructing networks from a dynamical system perspective for multivariate nonlinear time series.

Science.gov (United States)

Nakamura, Tomomichi; Tanizawa, Toshihiro; Small, Michael

2016-03-01

We describe a method for constructing networks for multivariate nonlinear time series. We approach the interaction between the various scalar time series from a deterministic dynamical system perspective and provide a generic and algorithmic test for whether the interaction between two measured time series is statistically significant. The method can be applied even when the data exhibit no obvious qualitative similarity: a situation in which the naive method utilizing the cross correlation function directly cannot correctly identify connectivity. To establish the connectivity between nodes we apply the previously proposed small-shuffle surrogate (SSS) method, which can investigate whether there are correlation structures in short-term variabilities (irregular fluctuations) between two data sets from the viewpoint of deterministic dynamical systems. The procedure to construct networks based on this idea is composed of three steps: (i) each time series is considered as a basic node of a network, (ii) the SSS method is applied to verify the connectivity between each pair of time series taken from the whole multivariate time series, and (iii) the pair of nodes is connected with an undirected edge when the null hypothesis cannot be rejected. The network constructed by the proposed method indicates the intrinsic (essential) connectivity of the elements included in the system or the underlying (assumed) system. The method is demonstrated for numerical data sets generated by known systems and applied to several experimental time series.

Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks.

Science.gov (United States)

Kim, D H; MacKinnon, T

2018-05-01

To identify the extent to which transfer learning from deep convolutional neural networks (CNNs), pre-trained on non-medical images, can be used for automated fracture detection on plain radiographs. The top layer of the Inception v3 network was re-trained using lateral wrist radiographs to produce a model for the classification of new studies as either "fracture" or "no fracture". The model was trained on a total of 11,112 images, after an eightfold data augmentation technique, from an initial set of 1,389 radiographs (695 "fracture" and 694 "no fracture"). The training data set was split 80:10:10 into training, validation, and test groups, respectively. An additional 100 wrist radiographs, comprising 50 "fracture" and 50 "no fracture" images, were used for final testing and statistical analysis. The area under the receiver operator characteristic curve (AUC) for this test was 0.954. Setting the diagnostic cut-off at a threshold designed to maximise both sensitivity and specificity resulted in values of 0.9 and 0.88, respectively. The AUC scores for this test were comparable to state-of-the-art providing proof of concept for transfer learning from CNNs in fracture detection on plain radiographs. This was achieved using only a moderate sample size. This technique is largely transferable, and therefore, has many potential applications in medical imaging, which may lead to significant improvements in workflow productivity and in clinical risk reduction. Copyright © 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.

Identification of nonlinear dynamics in power plant components using neural networks

International Nuclear Information System (INIS)

Parlos, A.G.; Fernandez, B.; Tsai, W.K.

1990-01-01

Advances in digital computer technology have enabled widespread implementation of closed-loop digital control systems in a variety of industries. In some instances, however, the complexity of the plant and the uncertainty associated with the parameters involved in the mathematical modeling narrow the range of applicability of most systematic control system design methodologies. A multiyear project has been initiated to assess the feasibility of the artificial neural networks (ANNs) technology for computerized enhanced diagnostics and control of nuclear power plant components. At this stage of the project, a new methodology, based on backpropagation learning, has been developed for identifying the nonlinear dynamic systems from a set of input-output data known as the training set

Deep nets vs expert designed features in medical physics: An IMRT QA case study.

Science.gov (United States)

Interian, Yannet; Rideout, Vincent; Kearney, Vasant P; Gennatas, Efstathios; Morin, Olivier; Cheung, Joey; Solberg, Timothy; Valdes, Gilmer

2018-03-30

The purpose of this study was to compare the performance of Deep Neural Networks against a technique designed by domain experts in the prediction of gamma passing rates for Intensity Modulated Radiation Therapy Quality Assurance (IMRT QA). A total of 498 IMRT plans across all treatment sites were planned in Eclipse version 11 and delivered using a dynamic sliding window technique on Clinac iX or TrueBeam Linacs. Measurements were performed using a commercial 2D diode array, and passing rates for 3%/3 mm local dose/distance-to-agreement (DTA) were recorded. Separately, fluence maps calculated for each plan were used as inputs to a convolution neural network (CNN). The CNNs were trained to predict IMRT QA gamma passing rates using TensorFlow and Keras. A set of model architectures, inspired by the convolutional blocks of the VGG-16 ImageNet model, were constructed and implemented. Synthetic data, created by rotating and translating the fluence maps during training, was created to boost the performance of the CNNs. Dropout, batch normalization, and data augmentation were utilized to help train the model. The performance of the CNNs was compared to a generalized Poisson regression model, previously developed for this application, which used 78 expert designed features. Deep Neural Networks without domain knowledge achieved comparable performance to a baseline system designed by domain experts in the prediction of 3%/3 mm Local gamma passing rates. An ensemble of neural nets resulted in a mean absolute error (MAE) of 0.70 ± 0.05 and the domain expert model resulted in a 0.74 ± 0.06. Convolutional neural networks (CNNs) with transfer learning can predict IMRT QA passing rates by automatically designing features from the fluence maps without human expert supervision. Predictions from CNNs are comparable to a system carefully designed by physicist experts. © 2018 American Association of Physicists in Medicine.

Generative Adversarial Networks for Noise Reduction in Low-Dose CT.

Science.gov (United States)

Wolterink, Jelmer M; Leiner, Tim; Viergever, Max A; Isgum, Ivana

2017-12-01

Noise is inherent to low-dose CT acquisition. We propose to train a convolutional neural network (CNN) jointly with an adversarial CNN to estimate routine-dose CT images from low-dose CT images and hence reduce noise. A generator CNN was trained to transform low-dose CT images into routine-dose CT images using voxelwise loss minimization. An adversarial discriminator CNN was simultaneously trained to distinguish the output of the generator from routine-dose CT images. The performance of this discriminator was used as an adversarial loss for the generator. Experiments were performed using CT images of an anthropomorphic phantom containing calcium inserts, as well as patient non-contrast-enhanced cardiac CT images. The phantom and patients were scanned at 20% and 100% routine clinical dose. Three training strategies were compared: the first used only voxelwise loss, the second combined voxelwise loss and adversarial loss, and the third used only adversarial loss. The results showed that training with only voxelwise loss resulted in the highest peak signal-to-noise ratio with respect to reference routine-dose images. However, CNNs trained with adversarial loss captured image statistics of routine-dose images better. Noise reduction improved quantification of low-density calcified inserts in phantom CT images and allowed coronary calcium scoring in low-dose patient CT images with high noise levels. Testing took less than 10 s per CT volume. CNN-based low-dose CT noise reduction in the image domain is feasible. Training with an adversarial network improves the CNNs ability to generate images with an appearance similar to that of reference routine-dose CT images.

The Use of Three-Dimensional Convolutional Neural Networks to Interpret LiDAR for Forest Inventory

Directory of Open Access Journals (Sweden)

Elias Ayrey

2018-04-01

Full Text Available As light detection and ranging (LiDAR technology becomes more available, it has become common to use these datasets to generate remotely sensed forest inventories across landscapes. Traditional methods for generating these inventories employ the use of height and proportion metrics to measure LiDAR returns and relate these back to field data using predictive models. Here, we employ a three-dimensional convolutional neural network (CNN, a deep learning technique that scans the LiDAR data and automatically generates useful features for predicting forest attributes. We test the accuracy in estimating forest attributes using the three-dimensional implementations of different CNN models commonly used in the field of image recognition. Using the best performing model architecture, we compared CNN performance to models developed using traditional height metrics. The results of this comparison show that CNNs produced 12% less prediction error when estimating biomass, 6% less in estimating tree count, and 2% less when estimating the percentage of needleleaf trees. We conclude that using CNNs can be a more accurate means of interpreting LiDAR data for forest inventories compared to standard approaches.

Competing edge networks

Science.gov (United States)

Parsons, Mark; Grindrod, Peter

2012-06-01

We introduce a model for a pair of nonlinear evolving networks, defined over a common set of vertices, subject to edgewise competition. Each network may grow new edges spontaneously or through triad closure. Both networks inhibit the other's growth and encourage the other's demise. These nonlinear stochastic competition equations yield to a mean field analysis resulting in a nonlinear deterministic system. There may be multiple equilibria; and bifurcations of different types are shown to occur within a reduced parameter space. This situation models competitive communication networks such as BlackBerry Messenger displacing SMS; or instant messaging displacing emails.

A fusion network for semantic segmentation using RGB-D data

Science.gov (United States)

Yuan, Jiahui; Zhang, Kun; Xia, Yifan; Qi, Lin; Dong, Junyu

2018-04-01

Semantic scene parsing is considerable in many intelligent field, including perceptual robotics. For the past few years, pixel-wise prediction tasks like semantic segmentation with RGB images has been extensively studied and has reached very remarkable parsing levels, thanks to convolutional neural networks (CNNs) and large scene datasets. With the development of stereo cameras and RGBD sensors, it is expected that additional depth information will help improving accuracy. In this paper, we propose a semantic segmentation framework incorporating RGB and complementary depth information. Motivated by the success of fully convolutional networks (FCN) in semantic segmentation field, we design a fully convolutional networks consists of two branches which extract features from both RGB and depth data simultaneously and fuse them as the network goes deeper. Instead of aggregating multiple model, our goal is to utilize RGB data and depth data more effectively in a single model. We evaluate our approach on the NYU-Depth V2 dataset, which consists of 1449 cluttered indoor scenes, and achieve competitive results with the state-of-the-art methods.

Nonlinear Dynamics on Interconnected Networks

Science.gov (United States)

Arenas, Alex; De Domenico, Manlio

2016-06-01

Networks of dynamical interacting units can represent many complex systems, from the human brain to transportation systems and societies. The study of these complex networks, when accounting for different types of interactions has become a subject of interest in the last few years, especially because its representational power in the description of users' interactions in diverse online social platforms (Facebook, Twitter, Instagram, etc.) [1], or in representing different transportation modes in urban networks [2,3]. The general name coined for these networks is multilayer networks, where each layer accounts for a type of interaction (see Fig. 1).

Decentralized Gauss-Newton method for nonlinear least squares on wide area network

Science.gov (United States)

Liu, Lanchao; Ling, Qing; Han, Zhu

2014-10-01

This paper presents a decentralized approach of Gauss-Newton (GN) method for nonlinear least squares (NLLS) on wide area network (WAN). In a multi-agent system, a centralized GN for NLLS requires the global GN Hessian matrix available at a central computing unit, which may incur large communication overhead. In the proposed decentralized alternative, each agent only needs local GN Hessian matrix to update iterates with the cooperation of neighbors. The detail formulation of decentralized NLLS on WAN is given, and the iteration at each agent is defined. The convergence property of the decentralized approach is analyzed, and numerical results validate the effectiveness of the proposed algorithm.

A comparative study between nonlinear regression and artificial neural network approaches for modelling wild oat (Avena fatua) field emergence

Science.gov (United States)

Non-linear regression techniques are used widely to fit weed field emergence patterns to soil microclimatic indices using S-type functions. Artificial neural networks present interesting and alternative features for such modeling purposes. In this work, a univariate hydrothermal-time based Weibull m...

A Decentralized Approach for Nonlinear Prediction of Time Series Data in Sensor Networks

Directory of Open Access Journals (Sweden)

Richard Cédric

2010-01-01

Full Text Available Wireless sensor networks rely on sensor devices deployed in an environment to support sensing and monitoring, including temperature, humidity, motion, and acoustic. Here, we propose a new approach to model physical phenomena and track their evolution by taking advantage of the recent developments of pattern recognition for nonlinear functional learning. These methods are, however, not suitable for distributed learning in sensor networks as the order of models scales linearly with the number of deployed sensors and measurements. In order to circumvent this drawback, we propose to design reduced order models by using an easy to compute sparsification criterion. We also propose a kernel-based least-mean-square algorithm for updating the model parameters using data collected by each sensor. The relevance of our approach is illustrated by two applications that consist of estimating a temperature distribution and tracking its evolution over time.

Deep convolutional neural networks as strong gravitational lens detectors

Science.gov (United States)

Schaefer, C.; Geiger, M.; Kuntzer, T.; Kneib, J.-P.

2018-03-01

Context. Future large-scale surveys with high-resolution imaging will provide us with approximately 105 new strong galaxy-scale lenses. These strong-lensing systems will be contained in large data amounts, however, which are beyond the capacity of human experts to visually classify in an unbiased way. Aim. We present a new strong gravitational lens finder based on convolutional neural networks (CNNs). The method was applied to the strong-lensing challenge organized by the Bologna Lens Factory. It achieved first and third place, respectively, on the space-based data set and the ground-based data set. The goal was to find a fully automated lens finder for ground-based and space-based surveys that minimizes human inspection. Methods: We compared the results of our CNN architecture and three new variations ("invariant" "views" and "residual") on the simulated data of the challenge. Each method was trained separately five times on 17 000 simulated images, cross-validated using 3000 images, and then applied to a test set with 100 000 images. We used two different metrics for evaluation, the area under the receiver operating characteristic curve (AUC) score, and the recall with no false positive (Recall0FP). Results: For ground-based data, our best method achieved an AUC score of 0.977 and a Recall0FP of 0.50. For space-based data, our best method achieved an AUC score of 0.940 and a Recall0FP of 0.32. Adding dihedral invariance to the CNN architecture diminished the overall score on space-based data, but achieved a higher no-contamination recall. We found that using committees of five CNNs produced the best recall at zero contamination and consistently scored better AUC than a single CNN. Conclusions: We found that for every variation of our CNN lensfinder, we achieved AUC scores close to 1 within 6%. A deeper network did not outperform simpler CNN models either. This indicates that more complex networks are not needed to model the simulated lenses. To verify this, more

Neural-network-observer-based optimal control for unknown nonlinear systems using adaptive dynamic programming

Science.gov (United States)

Liu, Derong; Huang, Yuzhu; Wang, Ding; Wei, Qinglai

2013-09-01

In this paper, an observer-based optimal control scheme is developed for unknown nonlinear systems using adaptive dynamic programming (ADP) algorithm. First, a neural-network (NN) observer is designed to estimate system states. Then, based on the observed states, a neuro-controller is constructed via ADP method to obtain the optimal control. In this design, two NN structures are used: a three-layer NN is used to construct the observer which can be applied to systems with higher degrees of nonlinearity and without a priori knowledge of system dynamics, and a critic NN is employed to approximate the value function. The optimal control law is computed using the critic NN and the observer NN. Uniform ultimate boundedness of the closed-loop system is guaranteed. The actor, critic, and observer structures are all implemented in real-time, continuously and simultaneously. Finally, simulation results are presented to demonstrate the effectiveness of the proposed control scheme.

Linear matrix inequality approach for synchronization control of fuzzy cellular neural networks with mixed time delays

International Nuclear Information System (INIS)

Balasubramaniam, P.; Kalpana, M.; Rakkiyappan, R.

2012-01-01

Fuzzy cellular neural networks (FCNNs) are special kinds of cellular neural networks (CNNs). Each cell in an FCNN contains fuzzy operating abilities. The entire network is governed by cellular computing laws. The design of FCNNs is based on fuzzy local rules. In this paper, a linear matrix inequality (LMI) approach for synchronization control of FCNNs with mixed delays is investigated. Mixed delays include discrete time-varying delays and unbounded distributed delays. A dynamic control scheme is proposed to achieve the synchronization between a drive network and a response network. By constructing the Lyapunov—Krasovskii functional which contains a triple-integral term and the free-weighting matrices method an improved delay-dependent stability criterion is derived in terms of LMIs. The controller can be easily obtained by solving the derived LMIs. A numerical example and its simulations are presented to illustrate the effectiveness of the proposed method. (interdisciplinary physics and related areas of science and technology)

The Role of Nonlinearity in Computing Graph-Theoretical Properties of Resting-State Functional Magnetic Resonance Imaging Brain Networks

Czech Academy of Sciences Publication Activity Database

Hartman, David; Hlinka, Jaroslav; Paluš, Milan; Mantini, D.; Corbetta, M.

2011-01-01

Roč. 21, č. 1 (2011), art.no 013119 ISSN 1054-1500 R&D Projects: GA MŠk 7E08027 Institutional research plan: CEZ:AV0Z10300504 Keywords : complex network * fMRI * brain connectivity * nonlinear * mutual information * correlation Subject RIV: BB - Applied Statistics, Operational Research Impact factor: 2.076, year: 2011

Anti-3D Weapon Model Detection for Safe 3D Printing Based on Convolutional Neural Networks and D2 Shape Distribution

Directory of Open Access Journals (Sweden)

Giao N. Pham

2018-03-01

Full Text Available With the development of 3D printing, weapons are easily printed without any restriction from the production managers. Therefore, anti-3D weapon model detection is necessary issue in safe 3D printing to prevent the printing of 3D weapon models. In this paper, we would like to propose an anti-3D weapon model detection algorithm to prevent the printing of anti-3D weapon models for safe 3D printing based on the D2 shape distribution and an improved convolutional neural networks (CNNs. The purpose of the proposed algorithm is to detect anti-3D weapon models when they are used in 3D printing. The D2 shape distribution is computed from random points on the surface of a 3D weapon model and their geometric features in order to construct a D2 vector. The D2 vector is then trained by improved CNNs. The CNNs are used to detect anti-3D weapon models for safe 3D printing by training D2 vectors which have been constructed from the D2 shape distribution of 3D weapon models. Experiments with 3D weapon models proved that the D2 shape distribution of 3D weapon models in the same class is the same. Training and testing results also verified that the accuracy of the proposed algorithm is higher than the conventional works. The proposed algorithm is applied in a small application, and it could detect anti-3D weapon models for safe 3D printing.

Thermodynamics of random reaction networks.

Science.gov (United States)

Fischer, Jakob; Kleidon, Axel; Dittrich, Peter

2015-01-01

Reaction networks are useful for analyzing reaction systems occurring in chemistry, systems biology, or Earth system science. Despite the importance of thermodynamic disequilibrium for many of those systems, the general thermodynamic properties of reaction networks are poorly understood. To circumvent the problem of sparse thermodynamic data, we generate artificial reaction networks and investigate their non-equilibrium steady state for various boundary fluxes. We generate linear and nonlinear networks using four different complex network models (Erdős-Rényi, Barabási-Albert, Watts-Strogatz, Pan-Sinha) and compare their topological properties with real reaction networks. For similar boundary conditions the steady state flow through the linear networks is about one order of magnitude higher than the flow through comparable nonlinear networks. In all networks, the flow decreases with the distance between the inflow and outflow boundary species, with Watts-Strogatz networks showing a significantly smaller slope compared to the three other network types. The distribution of entropy production of the individual reactions inside the network follows a power law in the intermediate region with an exponent of circa -1.5 for linear and -1.66 for nonlinear networks. An elevated entropy production rate is found in reactions associated with weakly connected species. This effect is stronger in nonlinear networks than in the linear ones. Increasing the flow through the nonlinear networks also increases the number of cycles and leads to a narrower distribution of chemical potentials. We conclude that the relation between distribution of dissipation, network topology and strength of disequilibrium is nontrivial and can be studied systematically by artificial reaction networks.

BrainNetCNN: Convolutional neural networks for brain networks; towards predicting neurodevelopment.

Science.gov (United States)

Kawahara, Jeremy; Brown, Colin J; Miller, Steven P; Booth, Brian G; Chau, Vann; Grunau, Ruth E; Zwicker, Jill G; Hamarneh, Ghassan

2017-02-01

We propose BrainNetCNN, a convolutional neural network (CNN) framework to predict clinical neurodevelopmental outcomes from brain networks. In contrast to the spatially local convolutions done in traditional image-based CNNs, our BrainNetCNN is composed of novel edge-to-edge, edge-to-node and node-to-graph convolutional filters that leverage the topological locality of structural brain networks. We apply the BrainNetCNN framework to predict cognitive and motor developmental outcome scores from structural brain networks of infants born preterm. Diffusion tensor images (DTI) of preterm infants, acquired between 27 and 46 weeks gestational age, were used to construct a dataset of structural brain connectivity networks. We first demonstrate the predictive capabilities of BrainNetCNN on synthetic phantom networks with simulated injury patterns and added noise. BrainNetCNN outperforms a fully connected neural-network with the same number of model parameters on both phantoms with focal and diffuse injury patterns. We then apply our method to the task of joint prediction of Bayley-III cognitive and motor scores, assessed at 18 months of age, adjusted for prematurity. We show that our BrainNetCNN framework outperforms a variety of other methods on the same data. Furthermore, BrainNetCNN is able to identify an infant's postmenstrual age to within about 2 weeks. Finally, we explore the high-level features learned by BrainNetCNN by visualizing the importance of each connection in the brain with respect to predicting the outcome scores. These findings are then discussed in the context of the anatomy and function of the developing preterm infant brain. Copyright © 2016 Elsevier Inc. All rights reserved.

TCSC Nonlinear Adaptive Damping Controller Design Based on RBF Neural Network to Enhance Power System Stability

DEFF Research Database (Denmark)

Yao, Wei; Fang, Jiakun; Zhao, Ping

2013-01-01

the characteristics of the conventional PID, but adjust the parameters of PID controller online using identified Jacobian information from RBFNN. Hence, it has strong adaptability to the variation of the system operating condition. The effectiveness of the proposed controller is tested on a two-machine five-bus power...... system and a four-machine two-area power system under different operating conditions in comparison with the lead-lag damping controller tuned by evolutionary algorithm (EA). Simulation results show that the proposed damping controller achieves good robust performance for damping the low frequency......In this paper, a nonlinear adaptive damping controller based on radial basis function neural network (RBFNN), which can infinitely approximate to nonlinear system, is proposed for thyristor controlled series capacitor (TCSC). The proposed TCSC adaptive damping controller can not only have...

Robust variable selection method for nonparametric differential equation models with application to nonlinear dynamic gene regulatory network analysis.

Science.gov (United States)

Lu, Tao

2016-01-01

The gene regulation network (GRN) evaluates the interactions between genes and look for models to describe the gene expression behavior. These models have many applications; for instance, by characterizing the gene expression mechanisms that cause certain disorders, it would be possible to target those genes to block the progress of the disease. Many biological processes are driven by nonlinear dynamic GRN. In this article, we propose a nonparametric differential equation (ODE) to model the nonlinear dynamic GRN. Specially, we address following questions simultaneously: (i) extract information from noisy time course gene expression data; (ii) model the nonlinear ODE through a nonparametric smoothing function; (iii) identify the important regulatory gene(s) through a group smoothly clipped absolute deviation (SCAD) approach; (iv) test the robustness of the model against possible shortening of experimental duration. We illustrate the usefulness of the model and associated statistical methods through a simulation and a real application examples.

Convolutional Neural Network for Multi-Source Deep Learning Crop Classification in Ukraine

Science.gov (United States)

Lavreniuk, M. S.

2016-12-01

Land cover and crop type maps are one of the most essential inputs when dealing with environmental and agriculture monitoring tasks [1]. During long time neural network (NN) approach was one of the most efficient and popular approach for most applications, including crop classification using remote sensing data, with high an overall accuracy (OA) [2]. In the last years the most popular and efficient method for multi-sensor and multi-temporal land cover classification is convolution neural networks (CNNs). Taking into account presence clouds in optical data, self-organizing Kohonen maps (SOMs) are used to restore missing pixel values in a time series of optical imagery from Landsat-8 satellite. After missing data restoration, optical data from Landsat-8 was merged with Sentinel-1A radar data for better crop types discrimination [3]. An ensemble of CNNs is proposed for multi-temporal satellite images supervised classification. Each CNN in the corresponding ensemble is a 1-d CNN with 4 layers implemented using the Google's library TensorFlow. The efficiency of the proposed approach was tested on a time-series of Landsat-8 and Sentinel-1A images over the JECAM test site (Kyiv region) in Ukraine in 2015. Overall classification accuracy for ensemble of CNNs was 93.5% that outperformed an ensemble of multi-layer perceptrons (MLPs) by +0.8% and allowed us to better discriminate summer crops, in particular maize and soybeans. For 2016 we would like to validate this method using Sentinel-1 and Sentinel-2 data for Ukraine territory within ESA project on country level demonstration Sen2Agri. 1. A. Kolotii et al., "Comparison of biophysical and satellite predictors for wheat yield forecasting in Ukraine," The Int. Arch. of Photogram., Rem. Sens. and Spatial Inform. Scie., vol. 40, no. 7, pp. 39-44, 2015. 2. F. Waldner et al., "Towards a set of agrosystem-specific cropland mapping methods to address the global cropland diversity," Int. Journal of Rem. Sens. vol. 37, no. 14, pp

Digital non-linear equalization for flexible capacity ultradense WDM channels for metro core networking

DEFF Research Database (Denmark)

Arlunno, Valeria; Zhang, Xu; Larsen, Knud J.

2011-01-01

carriers, we demonstrate that a digital non-linear equalization allow to mitigate inter-channel interference and improve overall system performance in terms of OSNR. Evaluation of the algorithm and comparison with an ultradense WDM system with coherent carriers generated from a single laser are also......An experimental demonstration of Ultradense WDM with advanced digital signal processing is presented. The scheme proposed allows the use of independent tunable DFB lasers spaced at 12.5 GHz for ultradense WDM PM-QPSK flexible capacity channels for metro core networking. To allocate extremely closed...

Recurrent Neural Networks to Correct Satellite Image Classification Maps

Science.gov (United States)

Maggiori, Emmanuel; Charpiat, Guillaume; Tarabalka, Yuliya; Alliez, Pierre

2017-09-01

While initially devised for image categorization, convolutional neural networks (CNNs) are being increasingly used for the pixelwise semantic labeling of images. However, the proper nature of the most common CNN architectures makes them good at recognizing but poor at localizing objects precisely. This problem is magnified in the context of aerial and satellite image labeling, where a spatially fine object outlining is of paramount importance. Different iterative enhancement algorithms have been presented in the literature to progressively improve the coarse CNN outputs, seeking to sharpen object boundaries around real image edges. However, one must carefully design, choose and tune such algorithms. Instead, our goal is to directly learn the iterative process itself. For this, we formulate a generic iterative enhancement process inspired from partial differential equations, and observe that it can be expressed as a recurrent neural network (RNN). Consequently, we train such a network from manually labeled data for our enhancement task. In a series of experiments we show that our RNN effectively learns an iterative process that significantly improves the quality of satellite image classification maps.

Noise-induced modulation of the relaxation kinetics around a non-equilibrium steady state of non-linear chemical reaction networks.

Science.gov (United States)

Ramaswamy, Rajesh; Sbalzarini, Ivo F; González-Segredo, Nélido

2011-01-28

Stochastic effects from correlated noise non-trivially modulate the kinetics of non-linear chemical reaction networks. This is especially important in systems where reactions are confined to small volumes and reactants are delivered in bursts. We characterise how the two noise sources confinement and burst modulate the relaxation kinetics of a non-linear reaction network around a non-equilibrium steady state. We find that the lifetimes of species change with burst input and confinement. Confinement increases the lifetimes of all species that are involved in any non-linear reaction as a reactant. Burst monotonically increases or decreases lifetimes. Competition between burst-induced and confinement-induced modulation may hence lead to a non-monotonic modulation. We quantify lifetime as the integral of the time autocorrelation function (ACF) of concentration fluctuations around a non-equilibrium steady state of the reaction network. Furthermore, we look at the first and second derivatives of the ACF, each of which is affected in opposite ways by burst and confinement. This allows discriminating between these two noise sources. We analytically derive the ACF from the linear Fokker-Planck approximation of the chemical master equation in order to establish a baseline for the burst-induced modulation at low confinement. Effects of higher confinement are then studied using a partial-propensity stochastic simulation algorithm. The results presented here may help understand the mechanisms that deviate stochastic kinetics from its deterministic counterpart. In addition, they may be instrumental when using fluorescence-lifetime imaging microscopy (FLIM) or fluorescence-correlation spectroscopy (FCS) to measure confinement and burst in systems with known reaction rates, or, alternatively, to correct for the effects of confinement and burst when experimentally measuring reaction rates.

PARTICLE SWARM OPTIMIZATION (PSO FOR TRAINING OPTIMIZATION ON CONVOLUTIONAL NEURAL NETWORK (CNN

Directory of Open Access Journals (Sweden)

Arie Rachmad Syulistyo

2016-02-01

Full Text Available Neural network attracts plenty of researchers lately. Substantial number of renowned universities have developed neural network for various both academically and industrially applications. Neural network shows considerable performance on various purposes. Nevertheless, for complex applications, neural network’s accuracy significantly deteriorates. To tackle the aforementioned drawback, lot of researches had been undertaken on the improvement of the standard neural network. One of the most promising modifications on standard neural network for complex applications is deep learning method. In this paper, we proposed the utilization of Particle Swarm Optimization (PSO in Convolutional Neural Networks (CNNs, which is one of the basic methods in deep learning. The use of PSO on the training process aims to optimize the results of the solution vectors on CNN in order to improve the recognition accuracy. The data used in this research is handwritten digit from MNIST. The experiments exhibited that the accuracy can be attained in 4 epoch is 95.08%. This result was better than the conventional CNN and DBN.  The execution time was also almost similar to the conventional CNN. Therefore, the proposed method was a promising method.

Thermodynamics of random reaction networks.

Directory of Open Access Journals (Sweden)

Jakob Fischer

Full Text Available Reaction networks are useful for analyzing reaction systems occurring in chemistry, systems biology, or Earth system science. Despite the importance of thermodynamic disequilibrium for many of those systems, the general thermodynamic properties of reaction networks are poorly understood. To circumvent the problem of sparse thermodynamic data, we generate artificial reaction networks and investigate their non-equilibrium steady state for various boundary fluxes. We generate linear and nonlinear networks using four different complex network models (Erdős-Rényi, Barabási-Albert, Watts-Strogatz, Pan-Sinha and compare their topological properties with real reaction networks. For similar boundary conditions the steady state flow through the linear networks is about one order of magnitude higher than the flow through comparable nonlinear networks. In all networks, the flow decreases with the distance between the inflow and outflow boundary species, with Watts-Strogatz networks showing a significantly smaller slope compared to the three other network types. The distribution of entropy production of the individual reactions inside the network follows a power law in the intermediate region with an exponent of circa -1.5 for linear and -1.66 for nonlinear networks. An elevated entropy production rate is found in reactions associated with weakly connected species. This effect is stronger in nonlinear networks than in the linear ones. Increasing the flow through the nonlinear networks also increases the number of cycles and leads to a narrower distribution of chemical potentials. We conclude that the relation between distribution of dissipation, network topology and strength of disequilibrium is nontrivial and can be studied systematically by artificial reaction networks.

Cellular nonlinear networks for strike-point localization at JET

International Nuclear Information System (INIS)

Arena, P.; Fortuna, L.; Bruno, M.; Vagliasindi, G.; Murari, A.; Andrew, P.; Mazzitelli, G.

2005-01-01

At JET, the potential of fast image processing for real-time purposes is thoroughly investigated. Particular attention is devoted to smart sensors based on system on chip technology. The data of the infrared cameras were processed with a chip implementing a cellular nonlinear network (CNN) structure so as to support and complement the magnetic diagnostics in the real-time localization of the strike-point position in the divertor. The circuit consists of two layers of complementary metal-oxide semiconductor components, the first being the sensor and the second implementing the actual CNN. This innovative hardware has made it possible to determine the position of the maximum thermal load with a time resolution of the order of 30 ms. Good congruency has been found with the measurement from the thermocouples in the divertor, proving the potential of the infrared data in locating the region of the maximum thermal load. The results are also confirmed by JET magnetic codes, both those used for the equilibrium reconstructions and those devoted to the identification of the plasma boundary

Convolutional Neural Networks for Human Activity Recognition Using Body-Worn Sensors

Directory of Open Access Journals (Sweden)

Fernando Moya Rueda

2018-05-01

Full Text Available Human activity recognition (HAR is a classification task for recognizing human movements. Methods of HAR are of great interest as they have become tools for measuring occurrences and durations of human actions, which are the basis of smart assistive technologies and manual processes analysis. Recently, deep neural networks have been deployed for HAR in the context of activities of daily living using multichannel time-series. These time-series are acquired from body-worn devices, which are composed of different types of sensors. The deep architectures process these measurements for finding basic and complex features in human corporal movements, and for classifying them into a set of human actions. As the devices are worn at different parts of the human body, we propose a novel deep neural network for HAR. This network handles sequence measurements from different body-worn devices separately. An evaluation of the architecture is performed on three datasets, the Oportunity, Pamap2, and an industrial dataset, outperforming the state-of-the-art. In addition, different network configurations will also be evaluated. We find that applying convolutions per sensor channel and per body-worn device improves the capabilities of convolutional neural network (CNNs.

A Nonlinear Autoregressive Exogenous (NARX Neural Network Model for the Prediction of the Daily Direct Solar Radiation

Directory of Open Access Journals (Sweden)

Zina Boussaada

2018-03-01

Full Text Available The solar photovoltaic (PV energy has an important place among the renewable energy sources. Therefore, several researchers have been interested by its modelling and its prediction, in order to improve the management of the electrical systems which include PV arrays. Among the existing techniques, artificial neural networks have proved their performance in the prediction of the solar radiation. However, the existing neural network models don’t satisfy the requirements of certain specific situations such as the one analyzed in this paper. The aim of this research work is to supply, with electricity, a race sailboat using exclusively renewable sources. The developed solution predicts the direct solar radiation on a horizontal surface. For that, a Nonlinear Autoregressive Exogenous (NARX neural network is used. All the specific conditions of the sailboat operation are taken into account. The results show that the best prediction performance is obtained when the training phase of the neural network is performed periodically.

Recurrent fuzzy neural network backstepping control for the prescribed output tracking performance of nonlinear dynamic systems.

Science.gov (United States)

Han, Seong-Ik; Lee, Jang-Myung

2014-01-01

This paper proposes a backstepping control system that uses a tracking error constraint and recurrent fuzzy neural networks (RFNNs) to achieve a prescribed tracking performance for a strict-feedback nonlinear dynamic system. A new constraint variable was defined to generate the virtual control that forces the tracking error to fall within prescribed boundaries. An adaptive RFNN was also used to obtain the required improvement on the approximation performances in order to avoid calculating the explosive number of terms generated by the recursive steps of traditional backstepping control. The boundedness and convergence of the closed-loop system was confirmed based on the Lyapunov stability theory. The prescribed performance of the proposed control scheme was validated by using it to control the prescribed error of a nonlinear system and a robot manipulator. © 2013 ISA. Published by Elsevier Ltd. All rights reserved.

Distributed Adaptive Neural Network Output Tracking of Leader-Following High-Order Stochastic Nonlinear Multiagent Systems With Unknown Dead-Zone Input.

Science.gov (United States)

Hua, Changchun; Zhang, Liuliu; Guan, Xinping

2017-01-01

This paper studies the problem of distributed output tracking consensus control for a class of high-order stochastic nonlinear multiagent systems with unknown nonlinear dead-zone under a directed graph topology. The adaptive neural networks are used to approximate the unknown nonlinear functions and a new inequality is used to deal with the completely unknown dead-zone input. Then, we design the controllers based on backstepping method and the dynamic surface control technique. It is strictly proved that the resulting closed-loop system is stable in probability in the sense of semiglobally uniform ultimate boundedness and the tracking errors between the leader and the followers approach to a small residual set based on Lyapunov stability theory. Finally, two simulation examples are presented to show the effectiveness and the advantages of the proposed techniques.

Classifying images using restricted Boltzmann machines and convolutional neural networks

Science.gov (United States)

Zhao, Zhijun; Xu, Tongde; Dai, Chenyu

2017-07-01

To improve the feature recognition ability of deep model transfer learning, we propose a hybrid deep transfer learning method for image classification based on restricted Boltzmann machines (RBM) and convolutional neural networks (CNNs). It integrates learning abilities of two models, which conducts subject classification by exacting structural higher-order statistics features of images. While the method transfers the trained convolutional neural networks to the target datasets, fully-connected layers can be replaced by restricted Boltzmann machine layers; then the restricted Boltzmann machine layers and Softmax classifier are retrained, and BP neural network can be used to fine-tuned the hybrid model. The restricted Boltzmann machine layers has not only fully integrated the whole feature maps, but also learns the statistical features of target datasets in the view of the biggest logarithmic likelihood, thus removing the effects caused by the content differences between datasets. The experimental results show that the proposed method has improved the accuracy of image classification, outperforming other methods on Pascal VOC2007 and Caltech101 datasets.

Cluster Synchronization of Diffusively Coupled Nonlinear Systems: A Contraction-Based Approach

Science.gov (United States)

Aminzare, Zahra; Dey, Biswadip; Davison, Elizabeth N.; Leonard, Naomi Ehrich

2018-04-01

Finding the conditions that foster synchronization in networked nonlinear systems is critical to understanding a wide range of biological and mechanical systems. However, the conditions proved in the literature for synchronization in nonlinear systems with linear coupling, such as has been used to model neuronal networks, are in general not strict enough to accurately determine the system behavior. We leverage contraction theory to derive new sufficient conditions for cluster synchronization in terms of the network structure, for a network where the intrinsic nonlinear dynamics of each node may differ. Our result requires that network connections satisfy a cluster-input-equivalence condition, and we explore the influence of this requirement on network dynamics. For application to networks of nodes with FitzHugh-Nagumo dynamics, we show that our new sufficient condition is tighter than those found in previous analyses that used smooth or nonsmooth Lyapunov functions. Improving the analytical conditions for when cluster synchronization will occur based on network configuration is a significant step toward facilitating understanding and control of complex networked systems.

Calculating electronic tunnel currents in networks of disordered irregularly shaped nanoparticles by mapping networks to arrays of parallel nonlinear resistors

Energy Technology Data Exchange (ETDEWEB)

Aghili Yajadda, Mir Massoud [CSIRO Manufacturing Flagship, P.O. Box 218, Lindfield NSW 2070 (Australia)

2014-10-21

We have shown both theoretically and experimentally that tunnel currents in networks of disordered irregularly shaped nanoparticles (NPs) can be calculated by considering the networks as arrays of parallel nonlinear resistors. Each resistor is described by a one-dimensional or a two-dimensional array of equal size nanoparticles that the tunnel junction gaps between nanoparticles in each resistor is assumed to be equal. The number of tunnel junctions between two contact electrodes and the tunnel junction gaps between nanoparticles are found to be functions of Coulomb blockade energies. In addition, the tunnel barriers between nanoparticles were considered to be tilted at high voltages. Furthermore, the role of thermal expansion coefficient of the tunnel junction gaps on the tunnel current is taken into account. The model calculations fit very well to the experimental data of a network of disordered gold nanoparticles, a forest of multi-wall carbon nanotubes, and a network of few-layer graphene nanoplates over a wide temperature range (5-300 K) at low and high DC bias voltages (0.001 mV–50 V). Our investigations indicate, although electron cotunneling in networks of disordered irregularly shaped NPs may occur, non-Arrhenius behavior at low temperatures cannot be described by the cotunneling model due to size distribution in the networks and irregular shape of nanoparticles. Non-Arrhenius behavior of the samples at zero bias voltage limit was attributed to the disorder in the samples. Unlike the electron cotunneling model, we found that the crossover from Arrhenius to non-Arrhenius behavior occurs at two temperatures, one at a high temperature and the other at a low temperature.

A self-adaption compensation control for hysteresis nonlinearity in piezo-actuated stages based on Pi-sigma fuzzy neural network

Science.gov (United States)

Xu, Rui; Zhou, Miaolei

2018-04-01

Piezo-actuated stages are widely applied in the high-precision positioning field nowadays. However, the inherent hysteresis nonlinearity in piezo-actuated stages greatly deteriorates the positioning accuracy of piezo-actuated stages. This paper first utilizes a nonlinear autoregressive moving average with exogenous inputs (NARMAX) model based on the Pi-sigma fuzzy neural network (PSFNN) to construct an online rate-dependent hysteresis model for describing the hysteresis nonlinearity in piezo-actuated stages. In order to improve the convergence rate of PSFNN and modeling precision, we adopt the gradient descent algorithm featuring three different learning factors to update the model parameters. The convergence of the NARMAX model based on the PSFNN is analyzed effectively. To ensure that the parameters can converge to the true values, the persistent excitation condition is considered. Then, a self-adaption compensation controller is designed for eliminating the hysteresis nonlinearity in piezo-actuated stages. A merit of the proposed controller is that it can directly eliminate the complex hysteresis nonlinearity in piezo-actuated stages without any inverse dynamic models. To demonstrate the effectiveness of the proposed model and control methods, a set of comparative experiments are performed on piezo-actuated stages. Experimental results show that the proposed modeling and control methods have excellent performance.

Nonlinear fiber optics

CERN Document Server

Agrawal, Govind

2012-01-01

Since the 4e appeared, a fast evolution of the field has occurred. The 5e of this classic work provides an up-to-date account of the nonlinear phenomena occurring inside optical fibers, the basis of all our telecommunications infastructure as well as being used in the medical field. Reflecting the big developments in research, this new edition includes major new content: slow light effects, which offers a reduction in noise and power consumption and more ordered network traffic-stimulated Brillouin scattering; vectorial treatment of highly nonlinear fibers; and a brand new chapter o

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

Science.gov (United States)

Chen, Fangyue; Chen, Guanrong Ron; He, Guolong; Xu, Xiubin; He, Qinbin

2009-10-01

Universal perceptron (UP), a generalization of Rosenblatt's perceptron, is considered in this paper, which is capable of implementing all Boolean functions (BFs). In the classification of BFs, there are: 1) linearly separable Boolean function (LSBF) class, 2) parity Boolean function (PBF) class, and 3) non-LSBF and non-PBF class. To implement these functions, UP takes different kinds of simple topological structures in which each contains at most one hidden layer along with the smallest possible number of hidden neurons. Inspired by the concept of DNA sequences in biological systems, a novel learning algorithm named DNA-like learning is developed, which is able to quickly train a network with any prescribed BF. The focus is on performing LSBF and PBF by a single-layer perceptron (SLP) with the new algorithm. Two criteria for LSBF and PBF are proposed, respectively, and a new measure for a BF, named nonlinearly separable degree (NLSD), is introduced. In the sense of this measure, the PBF is the most complex one. The new algorithm has many advantages including, in particular, fast running speed, good robustness, and no need of considering the convergence property. For example, the number of iterations and computations in implementing the basic 2-bit logic operations such as AND, OR, and XOR by using the new algorithm is far smaller than the ones needed by using other existing algorithms such as error-correction (EC) and backpropagation (BP) algorithms. Moreover, the synaptic weights and threshold values derived from UP can be directly used in designing of the template of cellular neural networks (CNNs), which has been considered as a new spatial-temporal sensory computing paradigm.

Noise-induced modulation of the relaxation kinetics around a non-equilibrium steady state of non-linear chemical reaction networks.

Directory of Open Access Journals (Sweden)

Rajesh Ramaswamy

2011-01-01

Full Text Available Stochastic effects from correlated noise non-trivially modulate the kinetics of non-linear chemical reaction networks. This is especially important in systems where reactions are confined to small volumes and reactants are delivered in bursts. We characterise how the two noise sources confinement and burst modulate the relaxation kinetics of a non-linear reaction network around a non-equilibrium steady state. We find that the lifetimes of species change with burst input and confinement. Confinement increases the lifetimes of all species that are involved in any non-linear reaction as a reactant. Burst monotonically increases or decreases lifetimes. Competition between burst-induced and confinement-induced modulation may hence lead to a non-monotonic modulation. We quantify lifetime as the integral of the time autocorrelation function (ACF of concentration fluctuations around a non-equilibrium steady state of the reaction network. Furthermore, we look at the first and second derivatives of the ACF, each of which is affected in opposite ways by burst and confinement. This allows discriminating between these two noise sources. We analytically derive the ACF from the linear Fokker-Planck approximation of the chemical master equation in order to establish a baseline for the burst-induced modulation at low confinement. Effects of higher confinement are then studied using a partial-propensity stochastic simulation algorithm. The results presented here may help understand the mechanisms that deviate stochastic kinetics from its deterministic counterpart. In addition, they may be instrumental when using fluorescence-lifetime imaging microscopy (FLIM or fluorescence-correlation spectroscopy (FCS to measure confinement and burst in systems with known reaction rates, or, alternatively, to correct for the effects of confinement and burst when experimentally measuring reaction rates.

Multi-stream CNN: Learning representations based on human-related regions for action recognition

NARCIS (Netherlands)

Tu, Zhigang; Xie, Wei; Qin, Qianqing; Poppe, R.W.; Veltkamp, R.C.; Li, Baoxin; Yuan, Junsong

2018-01-01

The most successful video-based human action recognition methods rely on feature representations extracted using Convolutional Neural Networks (CNNs). Inspired by the two-stream network (TS-Net), we propose a multi-stream Convolutional Neural Network (CNN) architecture to recognize human actions. We

Competing edge networks

International Nuclear Information System (INIS)

Parsons, Mark; Grindrod, Peter

2012-01-01

We introduce a model for a pair of nonlinear evolving networks, defined over a common set of vertices, subject to edgewise competition. Each network may grow new edges spontaneously or through triad closure. Both networks inhibit the other's growth and encourage the other's demise. These nonlinear stochastic competition equations yield to a mean field analysis resulting in a nonlinear deterministic system. There may be multiple equilibria; and bifurcations of different types are shown to occur within a reduced parameter space. This situation models competitive communication networks such as BlackBerry Messenger displacing SMS; or instant messaging displacing emails. -- Highlights: ► A model for edgewise-competing evolving network pairs is introduced. ► Defined competition equations yield to a mean field analysis. ► Multiple equilibrium states and different bifurcation types can occur. ► The system is sensitive to sparse initial conditions and near unstable equilibriums.

A Neural-Network-Based Nonlinear Adaptive State-Observer for Pressurized Water Reactors

Directory of Open Access Journals (Sweden)

Zhe Dong

2013-10-01

Full Text Available Although there have been some severe nuclear accidents such as Three Mile Island (USA, Chernobyl (Ukraine and Fukushima (Japan, nuclear fission energy is still a source of clean energy that can substitute for fossil fuels in a centralized way and in a great amount with commercial availability and economic competitiveness. Since the pressurized water reactor (PWR is the most widely used nuclear fission reactor, its safe, stable and efficient operation is meaningful to the current rebirth of the nuclear fission energy industry. Power-level regulation is an important technique which can deeply affect the operation stability and efficiency of PWRs. Compared with the classical power-level controllers, the advanced power-level regulators could strengthen both the closed-loop stability and control performance by feeding back the internal state-variables. However, not all of the internal state variables of a PWR can be obtained directly by measurements. To implement advanced PWR power-level control law, it is necessary to develop a state-observer to reconstruct the unmeasurable state-variables. Since a PWR is naturally a complex nonlinear system with parameters varying with power-level, fuel burnup, xenon isotope production, control rod worth and etc., it is meaningful to design a nonlinear observer for the PWR with adaptability to system uncertainties. Due to this and the strong learning capability of the multi-layer perceptron (MLP neural network, an MLP-based nonlinear adaptive observer is given for PWRs. Based upon Lyapunov stability theory, it is proved theoretically that this newly-built observer can provide bounded and convergent state-observation. This observer is then applied to the state-observation of a special PWR, i.e., the nuclear heating reactor (NHR, and numerical simulation results not only verify its feasibility but also give the relationship between the observation performance and observer parameters.

A Universal 3D Voxel Descriptor for Solid-State Material Informatics with Deep Convolutional Neural Networks.

Science.gov (United States)

Kajita, Seiji; Ohba, Nobuko; Jinnouchi, Ryosuke; Asahi, Ryoji

2017-12-05

Material informatics (MI) is a promising approach to liberate us from the time-consuming Edisonian (trial and error) process for material discoveries, driven by machine-learning algorithms. Several descriptors, which are encoded material features to feed computers, were proposed in the last few decades. Especially to solid systems, however, their insufficient representations of three dimensionality of field quantities such as electron distributions and local potentials have critically hindered broad and practical successes of the solid-state MI. We develop a simple, generic 3D voxel descriptor that compacts any field quantities, in such a suitable way to implement convolutional neural networks (CNNs). We examine the 3D voxel descriptor encoded from the electron distribution by a regression test with 680 oxides data. The present scheme outperforms other existing descriptors in the prediction of Hartree energies that are significantly relevant to the long-wavelength distribution of the valence electrons. The results indicate that this scheme can forecast any functionals of field quantities just by learning sufficient amount of data, if there is an explicit correlation between the target properties and field quantities. This 3D descriptor opens a way to import prominent CNNs-based algorithms of supervised, semi-supervised and reinforcement learnings into the solid-state MI.

Remote Sensing Scene Classification Based on Convolutional Neural Networks Pre-Trained Using Attention-Guided Sparse Filters

Directory of Open Access Journals (Sweden)

Jingbo Chen

2018-02-01

Full Text Available Semantic-level land-use scene classification is a challenging problem, in which deep learning methods, e.g., convolutional neural networks (CNNs, have shown remarkable capacity. However, a lack of sufficient labeled images has proved a hindrance to increasing the land-use scene classification accuracy of CNNs. Aiming at this problem, this paper proposes a CNN pre-training method under the guidance of a human visual attention mechanism. Specifically, a computational visual attention model is used to automatically extract salient regions in unlabeled images. Then, sparse filters are adopted to learn features from these salient regions, with the learnt parameters used to initialize the convolutional layers of the CNN. Finally, the CNN is further fine-tuned on labeled images. Experiments are performed on the UCMerced and AID datasets, which show that when combined with a demonstrative CNN, our method can achieve 2.24% higher accuracy than a plain CNN and can obtain an overall accuracy of 92.43% when combined with AlexNet. The results indicate that the proposed method can effectively improve CNN performance using easy-to-access unlabeled images and thus will enhance the performance of land-use scene classification especially when a large-scale labeled dataset is unavailable.

Partial synchronization in networks of non-linearly coupled oscillators: The Deserter Hubs Model

Energy Technology Data Exchange (ETDEWEB)

Freitas, Celso, E-mail: cbnfreitas@gmail.com; Macau, Elbert, E-mail: elbert.macau@inpe.br [Associate Laboratory for Computing and Applied Mathematics - LAC, Brazilian National Institute for Space Research - INPE (Brazil); Pikovsky, Arkady, E-mail: pikovsky@uni-potsdam.de [Department of Physics and Astronomy, University of Potsdam, Germany and Department of Control Theory, Nizhni Novgorod State University, Gagarin Av. 23, 606950, Nizhni Novgorod (Russian Federation)

2015-04-15

We study the Deserter Hubs Model: a Kuramoto-like model of coupled identical phase oscillators on a network, where attractive and repulsive couplings are balanced dynamically due to nonlinearity of interactions. Under weak force, an oscillator tends to follow the phase of its neighbors, but if an oscillator is compelled to follow its peers by a sufficient large number of cohesive neighbors, then it actually starts to act in the opposite manner, i.e., in anti-phase with the majority. Analytic results yield that if the repulsion parameter is small enough in comparison with the degree of the maximum hub, then the full synchronization state is locally stable. Numerical experiments are performed to explore the model beyond this threshold, where the overall cohesion is lost. We report in detail partially synchronous dynamical regimes, like stationary phase-locking, multistability, periodic and chaotic states. Via statistical analysis of different network organizations like tree, scale-free, and random ones, we found a measure allowing one to predict relative abundance of partially synchronous stationary states in comparison to time-dependent ones.

Learning Low Dimensional Convolutional Neural Networks for High-Resolution Remote Sensing Image Retrieval

Directory of Open Access Journals (Sweden)

Weixun Zhou

2017-05-01

Full Text Available Learning powerful feature representations for image retrieval has always been a challenging task in the field of remote sensing. Traditional methods focus on extracting low-level hand-crafted features which are not only time-consuming but also tend to achieve unsatisfactory performance due to the complexity of remote sensing images. In this paper, we investigate how to extract deep feature representations based on convolutional neural networks (CNNs for high-resolution remote sensing image retrieval (HRRSIR. To this end, several effective schemes are proposed to generate powerful feature representations for HRRSIR. In the first scheme, a CNN pre-trained on a different problem is treated as a feature extractor since there are no sufficiently-sized remote sensing datasets to train a CNN from scratch. In the second scheme, we investigate learning features that are specific to our problem by first fine-tuning the pre-trained CNN on a remote sensing dataset and then proposing a novel CNN architecture based on convolutional layers and a three-layer perceptron. The novel CNN has fewer parameters than the pre-trained and fine-tuned CNNs and can learn low dimensional features from limited labelled images. The schemes are evaluated on several challenging, publicly available datasets. The results indicate that the proposed schemes, particularly the novel CNN, achieve state-of-the-art performance.

International Conference on Applications in Nonlinear Dynamics

CERN Document Server

Longhini, Patrick; Palacios, Antonio

2017-01-01

This book presents collaborative research works carried out by experimentalists and theorists around the world in the field of nonlinear dynamical systems. It provides a forum for applications of nonlinear systems while solving practical problems in science and engineering. Topics include: Applied Nonlinear Optics, Sensor, Radar & Communication Signal Processing, Nano Devices, Nonlinear Biomedical Applications, Circuits & Systems, Coupled Nonlinear Oscillator, Precision Timing Devices, Networks, and other contemporary topics in the general field of Nonlinear Science. This book provides a comprehensive report of the various research projects presented at the International Conference on Applications in Nonlinear Dynamics (ICAND 2016) held in Denver, Colorado, 2016. It can be a valuable tool for scientists and engineering interested in connecting ideas and methods in nonlinear dynamics with actual design, fabrication and implementation of engineering applications or devices.

Nonlinear network model analysis of vibrational energy transfer and localisation in the Fenna-Matthews-Olson complex

Science.gov (United States)

Morgan, Sarah E.; Cole, Daniel J.; Chin, Alex W.

2016-11-01

Collective protein modes are expected to be important for facilitating energy transfer in the Fenna-Matthews-Olson (FMO) complex of photosynthetic green sulphur bacteria, however to date little work has focussed on the microscopic details of these vibrations. The nonlinear network model (NNM) provides a computationally inexpensive approach to studying vibrational modes at the microscopic level in large protein structures, whilst incorporating anharmonicity in the inter-residue interactions which can influence protein dynamics. We apply the NNM to the entire trimeric FMO complex and find evidence for the existence of nonlinear discrete breather modes. These modes tend to transfer energy to the highly connected core pigments, potentially opening up alternative excitation energy transfer routes through their influence on pigment properties. Incorporating localised modes based on these discrete breathers in the optical spectra calculations for FMO using ab initio site energies and excitonic couplings can substantially improve their agreement with experimental results.

Neural Generalized Predictive Control of a non-linear Process

DEFF Research Database (Denmark)

Sørensen, Paul Haase; Nørgård, Peter Magnus; Ravn, Ole

1998-01-01

The use of neural network in non-linear control is made difficult by the fact the stability and robustness is not guaranteed and that the implementation in real time is non-trivial. In this paper we introduce a predictive controller based on a neural network model which has promising stability qu...... detail and discuss the implementation difficulties. The neural generalized predictive controller is tested on a pneumatic servo sys-tem.......The use of neural network in non-linear control is made difficult by the fact the stability and robustness is not guaranteed and that the implementation in real time is non-trivial. In this paper we introduce a predictive controller based on a neural network model which has promising stability...... qualities. The controller is a non-linear version of the well-known generalized predictive controller developed in linear control theory. It involves minimization of a cost function which in the present case has to be done numerically. Therefore, we develop the numerical algorithms necessary in substantial...

Modelling computer networks

International Nuclear Information System (INIS)

Max, G

2011-01-01

Traffic models in computer networks can be described as a complicated system. These systems show non-linear features and to simulate behaviours of these systems are also difficult. Before implementing network equipments users wants to know capability of their computer network. They do not want the servers to be overloaded during temporary traffic peaks when more requests arrive than the server is designed for. As a starting point for our study a non-linear system model of network traffic is established to exam behaviour of the network planned. The paper presents setting up a non-linear simulation model that helps us to observe dataflow problems of the networks. This simple model captures the relationship between the competing traffic and the input and output dataflow. In this paper, we also focus on measuring the bottleneck of the network, which was defined as the difference between the link capacity and the competing traffic volume on the link that limits end-to-end throughput. We validate the model using measurements on a working network. The results show that the initial model estimates well main behaviours and critical parameters of the network. Based on this study, we propose to develop a new algorithm, which experimentally determines and predict the available parameters of the network modelled.

Nonlinear identification and control a neural network approach

CERN Document Server

Liu, G P

2001-01-01

The series Advances in Industrial Control aims to report and encourage technology transfer in control engineering. The rapid development of control technology has an impact on all areas of the control discipline. New theory, new controllers, actuators, sensors, new industrial processes, computer methods, new applications, new philosophies . . . , new challenges. Much of this development work resides in industrial reports, feasibility study papers and the reports of advanced collaborative projects. The series otTers an opportunity for researchers to present an extended exposition of such new work in all aspects of industrial control for wider and rapid dissemination. The time for nonlinear control to enter routine application seems to be approaching. Nonlinear control has had a long gestation period but much ofthe past has been concerned with methods that involve formal nonlinear functional model representations. It seems more likely that the breakthough will come through the use of other more flexible and ame...

Caffe con Troll: Shallow Ideas to Speed Up Deep Learning.

Science.gov (United States)

Hadjis, Stefan; Abuzaid, Firas; Zhang, Ce; Ré, Christopher

2015-01-01

We present Caffe con Troll (CcT), a fully compatible end-to-end version of the popular framework Caffe with rebuilt internals. We built CcT to examine the performance characteristics of training and deploying general-purpose convolutional neural networks across different hardware architectures. We find that, by employing standard batching optimizations for CPU training, we achieve a 4.5× throughput improvement over Caffe on popular networks like CaffeNet. Moreover, with these improvements, the end-to-end training time for CNNs is directly proportional to the FLOPS delivered by the CPU, which enables us to efficiently train hybrid CPU-GPU systems for CNNs.

Nonlinear Dynamics in Gene Regulation Promote Robustness and Evolvability of Gene Expression Levels.

Science.gov (United States)

Steinacher, Arno; Bates, Declan G; Akman, Ozgur E; Soyer, Orkun S

2016-01-01

Cellular phenotypes underpinned by regulatory networks need to respond to evolutionary pressures to allow adaptation, but at the same time be robust to perturbations. This creates a conflict in which mutations affecting regulatory networks must both generate variance but also be tolerated at the phenotype level. Here, we perform mathematical analyses and simulations of regulatory networks to better understand the potential trade-off between robustness and evolvability. Examining the phenotypic effects of mutations, we find an inverse correlation between robustness and evolvability that breaks only with nonlinearity in the network dynamics, through the creation of regions presenting sudden changes in phenotype with small changes in genotype. For genotypes embedding low levels of nonlinearity, robustness and evolvability correlate negatively and almost perfectly. By contrast, genotypes embedding nonlinear dynamics allow expression levels to be robust to small perturbations, while generating high diversity (evolvability) under larger perturbations. Thus, nonlinearity breaks the robustness-evolvability trade-off in gene expression levels by allowing disparate responses to different mutations. Using analytical derivations of robustness and system sensitivity, we show that these findings extend to a large class of gene regulatory network architectures and also hold for experimentally observed parameter regimes. Further, the effect of nonlinearity on the robustness-evolvability trade-off is ensured as long as key parameters of the system display specific relations irrespective of their absolute values. We find that within this parameter regime genotypes display low and noisy expression levels. Examining the phenotypic effects of mutations, we find an inverse correlation between robustness and evolvability that breaks only with nonlinearity in the network dynamics. Our results provide a possible solution to the robustness-evolvability trade-off, suggest an explanation for

Nonlinear Dynamics in Gene Regulation Promote Robustness and Evolvability of Gene Expression Levels.

Directory of Open Access Journals (Sweden)

Arno Steinacher

Full Text Available Cellular phenotypes underpinned by regulatory networks need to respond to evolutionary pressures to allow adaptation, but at the same time be robust to perturbations. This creates a conflict in which mutations affecting regulatory networks must both generate variance but also be tolerated at the phenotype level. Here, we perform mathematical analyses and simulations of regulatory networks to better understand the potential trade-off between robustness and evolvability. Examining the phenotypic effects of mutations, we find an inverse correlation between robustness and evolvability that breaks only with nonlinearity in the network dynamics, through the creation of regions presenting sudden changes in phenotype with small changes in genotype. For genotypes embedding low levels of nonlinearity, robustness and evolvability correlate negatively and almost perfectly. By contrast, genotypes embedding nonlinear dynamics allow expression levels to be robust to small perturbations, while generating high diversity (evolvability under larger perturbations. Thus, nonlinearity breaks the robustness-evolvability trade-off in gene expression levels by allowing disparate responses to different mutations. Using analytical derivations of robustness and system sensitivity, we show that these findings extend to a large class of gene regulatory network architectures and also hold for experimentally observed parameter regimes. Further, the effect of nonlinearity on the robustness-evolvability trade-off is ensured as long as key parameters of the system display specific relations irrespective of their absolute values. We find that within this parameter regime genotypes display low and noisy expression levels. Examining the phenotypic effects of mutations, we find an inverse correlation between robustness and evolvability that breaks only with nonlinearity in the network dynamics. Our results provide a possible solution to the robustness-evolvability trade-off, suggest

MORE: mixed optimization for reverse engineering--an application to modeling biological networks response via sparse systems of nonlinear differential equations.

Science.gov (United States)

Sambo, Francesco; de Oca, Marco A Montes; Di Camillo, Barbara; Toffolo, Gianna; Stützle, Thomas

2012-01-01

Reverse engineering is the problem of inferring the structure of a network of interactions between biological variables from a set of observations. In this paper, we propose an optimization algorithm, called MORE, for the reverse engineering of biological networks from time series data. The model inferred by MORE is a sparse system of nonlinear differential equations, complex enough to realistically describe the dynamics of a biological system. MORE tackles separately the discrete component of the problem, the determination of the biological network topology, and the continuous component of the problem, the strength of the interactions. This approach allows us both to enforce system sparsity, by globally constraining the number of edges, and to integrate a priori information about the structure of the underlying interaction network. Experimental results on simulated and real-world networks show that the mixed discrete/continuous optimization approach of MORE significantly outperforms standard continuous optimization and that MORE is competitive with the state of the art in terms of accuracy of the inferred networks.

Nonlinear Motion Tracking by Deep Learning Architecture

Science.gov (United States)

Verma, Arnav; Samaiya, Devesh; Gupta, Karunesh K.

2018-03-01

In the world of Artificial Intelligence, object motion tracking is one of the major problems. The extensive research is being carried out to track people in crowd. This paper presents a unique technique for nonlinear motion tracking in the absence of prior knowledge of nature of nonlinear path that the object being tracked may follow. We achieve this by first obtaining the centroid of the object and then using the centroid as the current example for a recurrent neural network trained using real-time recurrent learning. We have tweaked the standard algorithm slightly and have accumulated the gradient for few previous iterations instead of using just the current iteration as is the norm. We show that for a single object, such a recurrent neural network is highly capable of approximating the nonlinearity of its path.

Device Applications of Nonlinear Dynamics

CERN Document Server

Baglio, Salvatore

2006-01-01

This edited book is devoted specifically to the applications of complex nonlinear dynamic phenomena to real systems and device applications. While in the past decades there has been significant progress in the theory of nonlinear phenomena under an assortment of system boundary conditions and preparations, there exist comparatively few devices that actually take this rich behavior into account. "Device Applications of Nonlinear Dynamics" applies and exploits this knowledge to make devices which operate more efficiently and cheaply, while affording the promise of much better performance. Given the current explosion of ideas in areas as diverse as molecular motors, nonlinear filtering theory, noise-enhanced propagation, stochastic resonance and networked systems, the time is right to integrate the progress of complex systems research into real devices.

A Comparative Study of Neural Networks and Fuzzy Systems in Modeling of a Nonlinear Dynamic System

Directory of Open Access Journals (Sweden)

Metin Demirtas

2011-07-01

Full Text Available The aim of this paper is to compare the neural networks and fuzzy modeling approaches on a nonlinear system. We have taken Permanent Magnet Brushless Direct Current (PMBDC motor data and have generated models using both approaches. The predictive performance of both methods was compared on the data set for model configurations. The paper describes the results of these tests and discusses the effects of changing model parameters on predictive and practical performance. Modeling sensitivity was used to compare for two methods.

Nonlinear Modeling of the PEMFC Based On NNARX Approach

OpenAIRE

Shan-Jen Cheng; Te-Jen Chang; Kuang-Hsiung Tan; Shou-Ling Kuo

2015-01-01

Polymer Electrolyte Membrane Fuel Cell (PEMFC) is such a time-vary nonlinear dynamic system. The traditional linear modeling approach is hard to estimate structure correctly of PEMFC system. From this reason, this paper presents a nonlinear modeling of the PEMFC using Neural Network Auto-regressive model with eXogenous inputs (NNARX) approach. The multilayer perception (MLP) network is applied to evaluate the structure of the NNARX model of PEMFC. The validity and accurac...

Adversarial training and dilated convolutions for brain MRI segmentation

NARCIS (Netherlands)

Moeskops, P.; Veta, M.; Lafarge, M.W.; Eppenhof, K.A.J.; Pluim, J.P.W.

2017-01-01

Convolutional neural networks (CNNs) have been applied to various automatic image segmentation tasks in medical image analysis, including brain MRI segmentation. Generative adversarial networks have recently gained popularity because of their power in generating images that are difficult to

Simplified LQG Control with Neural Networks

DEFF Research Database (Denmark)

Sørensen, O.

1997-01-01

A new neural network application for non-linear state control is described. One neural network is modelled to form a Kalmann predictor and trained to act as an optimal state observer for a non-linear process. Another neural network is modelled to form a state controller and trained to produce...

Ship detection in optical remote sensing images based on deep convolutional neural networks

Science.gov (United States)

Yao, Yuan; Jiang, Zhiguo; Zhang, Haopeng; Zhao, Danpei; Cai, Bowen

2017-10-01

Automatic ship detection in optical remote sensing images has attracted wide attention for its broad applications. Major challenges for this task include the interference of cloud, wave, wake, and the high computational expenses. We propose a fast and robust ship detection algorithm to solve these issues. The framework for ship detection is designed based on deep convolutional neural networks (CNNs), which provide the accurate locations of ship targets in an efficient way. First, the deep CNN is designed to extract features. Then, a region proposal network (RPN) is applied to discriminate ship targets and regress the detection bounding boxes, in which the anchors are designed by intrinsic shape of ship targets. Experimental results on numerous panchromatic images demonstrate that, in comparison with other state-of-the-art ship detection methods, our method is more efficient and achieves higher detection accuracy and more precise bounding boxes in different complex backgrounds.

Ensemble Nonlinear Autoregressive Exogenous Artificial Neural Networks for Short-Term Wind Speed and Power Forecasting.

Science.gov (United States)

Men, Zhongxian; Yee, Eugene; Lien, Fue-Sang; Yang, Zhiling; Liu, Yongqian

2014-01-01

Short-term wind speed and wind power forecasts (for a 72 h period) are obtained using a nonlinear autoregressive exogenous artificial neural network (ANN) methodology which incorporates either numerical weather prediction or high-resolution computational fluid dynamics wind field information as an exogenous input. An ensemble approach is used to combine the predictions from many candidate ANNs in order to provide improved forecasts for wind speed and power, along with the associated uncertainties in these forecasts. More specifically, the ensemble ANN is used to quantify the uncertainties arising from the network weight initialization and from the unknown structure of the ANN. All members forming the ensemble of neural networks were trained using an efficient particle swarm optimization algorithm. The results of the proposed methodology are validated using wind speed and wind power data obtained from an operational wind farm located in Northern China. The assessment demonstrates that this methodology for wind speed and power forecasting generally provides an improvement in predictive skills when compared to the practice of using an "optimal" weight vector from a single ANN while providing additional information in the form of prediction uncertainty bounds.

Effects of Degree-Biased Transmission Rate and Nonlinear Infectivity on Rumor Spreading in Complex Social Networks

International Nuclear Information System (INIS)

Naimi, Y.; Roshani, F.

2010-12-01

We introduce the generalized rumor spreading model and analytically investigate the epidemic spreading for this model on scale-free networks. To generalize the standard rumor spreading model (rumor model in which each node's infectivity equals its degree and all links have a uniform connectivity strength), we introduce not only the infectivity function to determine the simultaneous contacts that a given node (individual) establishes to its connected neighbors but also the connectivity strength function (CSF) for the direct link between two connected nodes that lead to degree-biased propagation of rumors. In the case of nonlinear functions, the generalization enters the infectivity's exponent α and the CSF's exponent β into the analytical rumor model. We show that one can adjust the exponents α and β to control the epidemic threshold which is absent for the standard rumor spreading model. In addition, we obtain the critical threshold for the generalized model on the finite scale-free network and compare our results with the standard model on the same network. We show that the generalized model has a greater threshold than the standard model. (author)

Land Cover Classification via Multitemporal Spatial Data by Deep Recurrent Neural Networks

Science.gov (United States)

Ienco, Dino; Gaetano, Raffaele; Dupaquier, Claire; Maurel, Pierre

2017-10-01

Nowadays, modern earth observation programs produce huge volumes of satellite images time series (SITS) that can be useful to monitor geographical areas through time. How to efficiently analyze such kind of information is still an open question in the remote sensing field. Recently, deep learning methods proved suitable to deal with remote sensing data mainly for scene classification (i.e. Convolutional Neural Networks - CNNs - on single images) while only very few studies exist involving temporal deep learning approaches (i.e Recurrent Neural Networks - RNNs) to deal with remote sensing time series. In this letter we evaluate the ability of Recurrent Neural Networks, in particular the Long-Short Term Memory (LSTM) model, to perform land cover classification considering multi-temporal spatial data derived from a time series of satellite images. We carried out experiments on two different datasets considering both pixel-based and object-based classification. The obtained results show that Recurrent Neural Networks are competitive compared to state-of-the-art classifiers, and may outperform classical approaches in presence of low represented and/or highly mixed classes. We also show that using the alternative feature representation generated by LSTM can improve the performances of standard classifiers.

Generalizing a nonlinear geophysical flood theory to medium-sized river networks

Science.gov (United States)

Gupta, Vijay K.; Mantilla, Ricardo; Troutman, Brent M.; Dawdy, David; Krajewski, Witold F.

2010-01-01

The central hypothesis of a nonlinear geophysical flood theory postulates that, given space-time rainfall intensity for a rainfall-runoff event, solutions of coupled mass and momentum conservation differential equations governing runoff generation and transport in a self-similar river network produce spatial scaling, or a power law, relation between peak discharge and drainage area in the limit of large area. The excellent fit of a power law for the destructive flood event of June 2008 in the 32,400-km2 Iowa River basin over four orders of magnitude variation in drainage areas supports the central hypothesis. The challenge of predicting observed scaling exponent and intercept from physical processes is explained. We show scaling in mean annual peak discharges, and briefly discuss that it is physically connected with scaling in multiple rainfall-runoff events. Scaling in peak discharges would hold in a non-stationary climate due to global warming but its slope and intercept would change.

Biological Networks Entropies: Examples in Neural Memory Networks, Genetic Regulation Networks and Social Epidemic Networks

Directory of Open Access Journals (Sweden)

Jacques Demongeot

2018-01-01

Full Text Available Networks used in biological applications at different scales (molecule, cell and population are of different types: neuronal, genetic, and social, but they share the same dynamical concepts, in their continuous differential versions (e.g., non-linear Wilson-Cowan system as well as in their discrete Boolean versions (e.g., non-linear Hopfield system; in both cases, the notion of interaction graph G(J associated to its Jacobian matrix J, and also the concepts of frustrated nodes, positive or negative circuits of G(J, kinetic energy, entropy, attractors, structural stability, etc., are relevant and useful for studying the dynamics and the robustness of these systems. We will give some general results available for both continuous and discrete biological networks, and then study some specific applications of three new notions of entropy: (i attractor entropy, (ii isochronal entropy and (iii entropy centrality; in three domains: a neural network involved in the memory evocation, a genetic network responsible of the iron control and a social network accounting for the obesity spread in high school environment.

Recurrent-Neural-Network-Based Multivariable Adaptive Control for a Class of Nonlinear Dynamic Systems With Time-Varying Delay.

Science.gov (United States)

Hwang, Chih-Lyang; Jan, Chau

2016-02-01

At the beginning, an approximate nonlinear autoregressive moving average (NARMA) model is employed to represent a class of multivariable nonlinear dynamic systems with time-varying delay. It is known that the disadvantages of robust control for the NARMA model are as follows: 1) suitable control parameters for larger time delay are more sensitive to achieving desirable performance; 2) it only deals with bounded uncertainty; and 3) the nominal NARMA model must be learned in advance. Due to the dynamic feature of the NARMA model, a recurrent neural network (RNN) is online applied to learn it. However, the system performance becomes deteriorated due to the poor learning of the larger variation of system vector functions. In this situation, a simple network is employed to compensate the upper bound of the residue caused by the linear parameterization of the approximation error of RNN. An e -modification learning law with a projection for weight matrix is applied to guarantee its boundedness without persistent excitation. Under suitable conditions, the semiglobally ultimately bounded tracking with the boundedness of estimated weight matrix is obtained by the proposed RNN-based multivariable adaptive control. Finally, simulations are presented to verify the effectiveness and robustness of the proposed control.

Neural Networks for Optimal Control

DEFF Research Database (Denmark)

Sørensen, O.

1995-01-01

Two neural networks are trained to act as an observer and a controller, respectively, to control a non-linear, multi-variable process.......Two neural networks are trained to act as an observer and a controller, respectively, to control a non-linear, multi-variable process....

Automated segmentation of geographic atrophy using deep convolutional neural networks

Science.gov (United States)

Hu, Zhihong; Wang, Ziyuan; Sadda, SriniVas R.

2018-02-01

Geographic atrophy (GA) is an end-stage manifestation of the advanced age-related macular degeneration (AMD), the leading cause of blindness and visual impairment in developed nations. Techniques to rapidly and precisely detect and quantify GA would appear to be of critical importance in advancing the understanding of its pathogenesis. In this study, we develop an automated supervised classification system using deep convolutional neural networks (CNNs) for segmenting GA in fundus autofluorescene (FAF) images. More specifically, to enhance the contrast of GA relative to the background, we apply the contrast limited adaptive histogram equalization. Blood vessels may cause GA segmentation errors due to similar intensity level to GA. A tensor-voting technique is performed to identify the blood vessels and a vessel inpainting technique is applied to suppress the GA segmentation errors due to the blood vessels. To handle the large variation of GA lesion sizes, three deep CNNs with three varying sized input image patches are applied. Fifty randomly chosen FAF images are obtained from fifty subjects with GA. The algorithm-defined GA regions are compared with manual delineation by a certified grader. A two-fold cross-validation is applied to evaluate the algorithm performance. The mean segmentation accuracy, true positive rate (i.e. sensitivity), true negative rate (i.e. specificity), positive predictive value, false discovery rate, and overlap ratio, between the algorithm- and manually-defined GA regions are 0.97 +/- 0.02, 0.89 +/- 0.08, 0.98 +/- 0.02, 0.87 +/- 0.12, 0.13 +/- 0.12, and 0.79 +/- 0.12 respectively, demonstrating a high level of agreement.

A single network adaptive critic (SNAC) architecture for optimal control synthesis for a class of nonlinear systems.

Science.gov (United States)

Padhi, Radhakant; Unnikrishnan, Nishant; Wang, Xiaohua; Balakrishnan, S N

2006-12-01

Even though dynamic programming offers an optimal control solution in a state feedback form, the method is overwhelmed by computational and storage requirements. Approximate dynamic programming implemented with an Adaptive Critic (AC) neural network structure has evolved as a powerful alternative technique that obviates the need for excessive computations and storage requirements in solving optimal control problems. In this paper, an improvement to the AC architecture, called the "Single Network Adaptive Critic (SNAC)" is presented. This approach is applicable to a wide class of nonlinear systems where the optimal control (stationary) equation can be explicitly expressed in terms of the state and costate variables. The selection of this terminology is guided by the fact that it eliminates the use of one neural network (namely the action network) that is part of a typical dual network AC setup. As a consequence, the SNAC architecture offers three potential advantages: a simpler architecture, lesser computational load and elimination of the approximation error associated with the eliminated network. In order to demonstrate these benefits and the control synthesis technique using SNAC, two problems have been solved with the AC and SNAC approaches and their computational performances are compared. One of these problems is a real-life Micro-Electro-Mechanical-system (MEMS) problem, which demonstrates that the SNAC technique is applicable to complex engineering systems.

Mathematical modeling and applications in nonlinear dynamics

CERN Document Server

Merdan, Hüseyin

2016-01-01

The book covers nonlinear physical problems and mathematical modeling, including molecular biology, genetics, neurosciences, artificial intelligence with classical problems in mechanics and astronomy and physics. The chapters present nonlinear mathematical modeling in life science and physics through nonlinear differential equations, nonlinear discrete equations and hybrid equations. Such modeling can be effectively applied to the wide spectrum of nonlinear physical problems, including the KAM (Kolmogorov-Arnold-Moser (KAM)) theory, singular differential equations, impulsive dichotomous linear systems, analytical bifurcation trees of periodic motions, and almost or pseudo- almost periodic solutions in nonlinear dynamical systems. Provides methods for mathematical models with switching, thresholds, and impulses, each of particular importance for discontinuous processes Includes qualitative analysis of behaviors on Tumor-Immune Systems and methods of analysis for DNA, neural networks and epidemiology Introduces...

Convolutional neural network-based classification system design with compressed wireless sensor network images.

Science.gov (United States)

Ahn, Jungmo; Park, JaeYeon; Park, Donghwan; Paek, Jeongyeup; Ko, JeongGil

2018-01-01

With the introduction of various advanced deep learning algorithms, initiatives for image classification systems have transitioned over from traditional machine learning algorithms (e.g., SVM) to Convolutional Neural Networks (CNNs) using deep learning software tools. A prerequisite in applying CNN to real world applications is a system that collects meaningful and useful data. For such purposes, Wireless Image Sensor Networks (WISNs), that are capable of monitoring natural environment phenomena using tiny and low-power cameras on resource-limited embedded devices, can be considered as an effective means of data collection. However, with limited battery resources, sending high-resolution raw images to the backend server is a burdensome task that has direct impact on network lifetime. To address this problem, we propose an energy-efficient pre- and post- processing mechanism using image resizing and color quantization that can significantly reduce the amount of data transferred while maintaining the classification accuracy in the CNN at the backend server. We show that, if well designed, an image in its highly compressed form can be well-classified with a CNN model trained in advance using adequately compressed data. Our evaluation using a real image dataset shows that an embedded device can reduce the amount of transmitted data by ∼71% while maintaining a classification accuracy of ∼98%. Under the same conditions, this process naturally reduces energy consumption by ∼71% compared to a WISN that sends the original uncompressed images.

Nonlinear dynamic systems identification using recurrent interval type-2 TSK fuzzy neural network - A novel structure.

Science.gov (United States)

El-Nagar, Ahmad M

2018-01-01

In this study, a novel structure of a recurrent interval type-2 Takagi-Sugeno-Kang (TSK) fuzzy neural network (FNN) is introduced for nonlinear dynamic and time-varying systems identification. It combines the type-2 fuzzy sets (T2FSs) and a recurrent FNN to avoid the data uncertainties. The fuzzy firing strengths in the proposed structure are returned to the network input as internal variables. The interval type-2 fuzzy sets (IT2FSs) is used to describe the antecedent part for each rule while the consequent part is a TSK-type, which is a linear function of the internal variables and the external inputs with interval weights. All the type-2 fuzzy rules for the proposed RIT2TSKFNN are learned on-line based on structure and parameter learning, which are performed using the type-2 fuzzy clustering. The antecedent and consequent parameters of the proposed RIT2TSKFNN are updated based on the Lyapunov function to achieve network stability. The obtained results indicate that our proposed network has a small root mean square error (RMSE) and a small integral of square error (ISE) with a small number of rules and a small computation time compared with other type-2 FNNs. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.

Short-Term Forecasting of Electric Loads Using Nonlinear Autoregressive Artificial Neural Networks with Exogenous Vector Inputs

Directory of Open Access Journals (Sweden)

Jaime Buitrago

2017-01-01

Full Text Available Short-term load forecasting is crucial for the operations planning of an electrical grid. Forecasting the next 24 h of electrical load in a grid allows operators to plan and optimize their resources. The purpose of this study is to develop a more accurate short-term load forecasting method utilizing non-linear autoregressive artificial neural networks (ANN with exogenous multi-variable input (NARX. The proposed implementation of the network is new: the neural network is trained in open-loop using actual load and weather data, and then, the network is placed in closed-loop to generate a forecast using the predicted load as the feedback input. Unlike the existing short-term load forecasting methods using ANNs, the proposed method uses its own output as the input in order to improve the accuracy, thus effectively implementing a feedback loop for the load, making it less dependent on external data. Using the proposed framework, mean absolute percent errors in the forecast in the order of 1% have been achieved, which is a 30% improvement on the average error using feedforward ANNs, ARMAX and state space methods, which can result in large savings by avoiding commissioning of unnecessary power plants. The New England electrical load data are used to train and validate the forecast prediction.

Global robust stability for shunting inhibitory CNNs with delays.

Science.gov (United States)

Wang, Lingna; Lin, Yiping

2004-08-01

In this paper, the problem of global robust stability for shunting inhibitory cellular neural networks (SICNNs) is studied. A sufficient condition guaranteeing the network's global robust stability is established. The result can easily be used to verify globally robust stable networks. An example is given to illustrate that the conditions of our results are feasible.

Conservation Laws in Biochemical Reaction Networks

DEFF Research Database (Denmark)

Mahdi, Adam; Ferragut, Antoni; Valls, Claudia

2017-01-01

We study the existence of linear and nonlinear conservation laws in biochemical reaction networks with mass-action kinetics. It is straightforward to compute the linear conservation laws as they are related to the left null-space of the stoichiometry matrix. The nonlinear conservation laws...... are difficult to identify and have rarely been considered in the context of mass-action reaction networks. Here, using the Darboux theory of integrability, we provide necessary structural (i.e., parameterindependent) conditions on a reaction network to guarantee the existence of nonlinear conservation laws...

Morphological neural networks

Energy Technology Data Exchange (ETDEWEB)

Ritter, G.X.; Sussner, P. [Univ. of Florida, Gainesville, FL (United States)

1996-12-31

The theory of artificial neural networks has been successfully applied to a wide variety of pattern recognition problems. In this theory, the first step in computing the next state of a neuron or in performing the next layer neural network computation involves the linear operation of multiplying neural values by their synaptic strengths and adding the results. Thresholding usually follows the linear operation in order to provide for nonlinearity of the network. In this paper we introduce a novel class of neural networks, called morphological neural networks, in which the operations of multiplication and addition are replaced by addition and maximum (or minimum), respectively. By taking the maximum (or minimum) of sums instead of the sum of products, morphological network computation is nonlinear before thresholding. As a consequence, the properties of morphological neural networks are drastically different than those of traditional neural network models. In this paper we consider some of these differences and provide some particular examples of morphological neural network.

A Novel Modeling Method for Aircraft Engine Using Nonlinear Autoregressive Exogenous (NARX) Models Based on Wavelet Neural Networks

Science.gov (United States)

Yu, Bing; Shu, Wenjun; Cao, Can

2018-05-01

A novel modeling method for aircraft engine using nonlinear autoregressive exogenous (NARX) models based on wavelet neural networks is proposed. The identification principle and process based on wavelet neural networks are studied, and the modeling scheme based on NARX is proposed. Then, the time series data sets from three types of aircraft engines are utilized to build the corresponding NARX models, and these NARX models are validated by the simulation. The results show that all the best NARX models can capture the original aircraft engine's dynamic characteristic well with the high accuracy. For every type of engine, the relative identification errors of its best NARX model and the component level model are no more than 3.5 % and most of them are within 1 %.

On the holistic approach in cellular and cancer biology: nonlinearity, complexity, and quasi-determinism of the dynamic cellular network.

Science.gov (United States)

Waliszewski, P; Molski, M; Konarski, J

1998-06-01

A keystone of the molecular reductionist approach to cellular biology is a specific deductive strategy relating genotype to phenotype-two distinct categories. This relationship is based on the assumption that the intermediary cellular network of actively transcribed genes and their regulatory elements is deterministic (i.e., a link between expression of a gene and a phenotypic trait can always be identified, and evolution of the network in time is predetermined). However, experimental data suggest that the relationship between genotype and phenotype is nonbijective (i.e., a gene can contribute to the emergence of more than just one phenotypic trait or a phenotypic trait can be determined by expression of several genes). This implies nonlinearity (i.e., lack of the proportional relationship between input and the outcome), complexity (i.e. emergence of the hierarchical network of multiple cross-interacting elements that is sensitive to initial conditions, possesses multiple equilibria, organizes spontaneously into different morphological patterns, and is controlled in dispersed rather than centralized manner), and quasi-determinism (i.e., coexistence of deterministic and nondeterministic events) of the network. Nonlinearity within the space of the cellular molecular events underlies the existence of a fractal structure within a number of metabolic processes, and patterns of tissue growth, which is measured experimentally as a fractal dimension. Because of its complexity, the same phenotype can be associated with a number of alternative sequences of cellular events. Moreover, the primary cause initiating phenotypic evolution of cells such as malignant transformation can be favored probabilistically, but not identified unequivocally. Thermodynamic fluctuations of energy rather than gene mutations, the material traits of the fluctuations alter both the molecular and informational structure of the network. Then, the interplay between deterministic chaos, complexity, self

Synchronization of chaotic systems and identification of nonlinear systems by using recurrent hierarchical type-2 fuzzy neural networks.

Science.gov (United States)

Mohammadzadeh, Ardashir; Ghaemi, Sehraneh

2015-09-01

This paper proposes a novel approach for training of proposed recurrent hierarchical interval type-2 fuzzy neural networks (RHT2FNN) based on the square-root cubature Kalman filters (SCKF). The SCKF algorithm is used to adjust the premise part of the type-2 FNN and the weights of defuzzification and the feedback weights. The recurrence property in the proposed network is the output feeding of each membership function to itself. The proposed RHT2FNN is employed in the sliding mode control scheme for the synchronization of chaotic systems. Unknown functions in the sliding mode control approach are estimated by RHT2FNN. Another application of the proposed RHT2FNN is the identification of dynamic nonlinear systems. The effectiveness of the proposed network and its learning algorithm is verified by several simulation examples. Furthermore, the universal approximation of RHT2FNNs is also shown. Copyright © 2015 ISA. Published by Elsevier Ltd. All rights reserved.

Synchronization, non-linear dynamics and low-frequency fluctuations: Analogy between spontaneous brain activity and networked single-transistor chaotic oscillators

International Nuclear Information System (INIS)

Minati, Ludovico; Chiesa, Pietro; Tabarelli, Davide; Jovicich, Jorge; D'Incerti, Ludovico

2015-01-01

In this paper, the topographical relationship between functional connectivity (intended as inter-regional synchronization), spectral and non-linear dynamical properties across cortical areas of the healthy human brain is considered. Based upon functional MRI acquisitions of spontaneous activity during wakeful idleness, node degree maps are determined by thresholding the temporal correlation coefficient among all voxel pairs. In addition, for individual voxel time-series, the relative amplitude of low-frequency fluctuations and the correlation dimension (D 2 ), determined with respect to Fourier amplitude and value distribution matched surrogate data, are measured. Across cortical areas, high node degree is associated with a shift towards lower frequency activity and, compared to surrogate data, clearer saturation to a lower correlation dimension, suggesting presence of non-linear structure. An attempt to recapitulate this relationship in a network of single-transistor oscillators is made, based on a diffusive ring (n = 90) with added long-distance links defining four extended hub regions. Similarly to the brain data, it is found that oscillators in the hub regions generate signals with larger low-frequency cycle amplitude fluctuations and clearer saturation to a lower correlation dimension compared to surrogates. The effect emerges more markedly close to criticality. The homology observed between the two systems despite profound differences in scale, coupling mechanism and dynamics appears noteworthy. These experimental results motivate further investigation into the heterogeneity of cortical non-linear dynamics in relation to connectivity and underline the ability for small networks of single-transistor oscillators to recreate collective phenomena arising in much more complex biological systems, potentially representing a future platform for modelling disease-related changes

Synchronization, non-linear dynamics and low-frequency fluctuations: Analogy between spontaneous brain activity and networked single-transistor chaotic oscillators

Energy Technology Data Exchange (ETDEWEB)

Minati, Ludovico, E-mail: lminati@ieee.org, E-mail: ludovico.minati@unitn.it, E-mail: lminati@istituto-besta.it [Scientific Department, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan (Italy); Center for Mind/Brain Sciences, University of Trento, Trento (Italy); Chiesa, Pietro; Tabarelli, Davide; Jovicich, Jorge [Center for Mind/Brain Sciences, University of Trento, Trento (Italy); D' Incerti, Ludovico [Neuroradiology Unit, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milan (Italy)

2015-03-15

In this paper, the topographical relationship between functional connectivity (intended as inter-regional synchronization), spectral and non-linear dynamical properties across cortical areas of the healthy human brain is considered. Based upon functional MRI acquisitions of spontaneous activity during wakeful idleness, node degree maps are determined by thresholding the temporal correlation coefficient among all voxel pairs. In addition, for individual voxel time-series, the relative amplitude of low-frequency fluctuations and the correlation dimension (D{sub 2}), determined with respect to Fourier amplitude and value distribution matched surrogate data, are measured. Across cortical areas, high node degree is associated with a shift towards lower frequency activity and, compared to surrogate data, clearer saturation to a lower correlation dimension, suggesting presence of non-linear structure. An attempt to recapitulate this relationship in a network of single-transistor oscillators is made, based on a diffusive ring (n = 90) with added long-distance links defining four extended hub regions. Similarly to the brain data, it is found that oscillators in the hub regions generate signals with larger low-frequency cycle amplitude fluctuations and clearer saturation to a lower correlation dimension compared to surrogates. The effect emerges more markedly close to criticality. The homology observed between the two systems despite profound differences in scale, coupling mechanism and dynamics appears noteworthy. These experimental results motivate further investigation into the heterogeneity of cortical non-linear dynamics in relation to connectivity and underline the ability for small networks of single-transistor oscillators to recreate collective phenomena arising in much more complex biological systems, potentially representing a future platform for modelling disease-related changes.

Robust synchronization analysis in nonlinear stochastic cellular networks with time-varying delays, intracellular perturbations and intercellular noise.

Science.gov (United States)

Chen, Po-Wei; Chen, Bor-Sen

2011-08-01

Naturally, a cellular network consisted of a large amount of interacting cells is complex. These cells have to be synchronized in order to emerge their phenomena for some biological purposes. However, the inherently stochastic intra and intercellular interactions are noisy and delayed from biochemical processes. In this study, a robust synchronization scheme is proposed for a nonlinear stochastic time-delay coupled cellular network (TdCCN) in spite of the time-varying process delay and intracellular parameter perturbations. Furthermore, a nonlinear stochastic noise filtering ability is also investigated for this synchronized TdCCN against stochastic intercellular and environmental disturbances. Since it is very difficult to solve a robust synchronization problem with the Hamilton-Jacobi inequality (HJI) matrix, a linear matrix inequality (LMI) is employed to solve this problem via the help of a global linearization method. Through this robust synchronization analysis, we can gain a more systemic insight into not only the robust synchronizability but also the noise filtering ability of TdCCN under time-varying process delays, intracellular perturbations and intercellular disturbances. The measures of robustness and noise filtering ability of a synchronized TdCCN have potential application to the designs of neuron transmitters, on-time mass production of biochemical molecules, and synthetic biology. Finally, a benchmark of robust synchronization design in Escherichia coli repressilators is given to confirm the effectiveness of the proposed methods. Copyright © 2011 Elsevier Inc. All rights reserved.

Ranking scientific publications: the effect of nonlinearity

Science.gov (United States)

Yao, Liyang; Wei, Tian; Zeng, An; Fan, Ying; di, Zengru

2014-10-01

Ranking the significance of scientific publications is a long-standing challenge. The network-based analysis is a natural and common approach for evaluating the scientific credit of papers. Although the number of citations has been widely used as a metric to rank papers, recently some iterative processes such as the well-known PageRank algorithm have been applied to the citation networks to address this problem. In this paper, we introduce nonlinearity to the PageRank algorithm when aggregating resources from different nodes to further enhance the effect of important papers. The validation of our method is performed on the data of American Physical Society (APS) journals. The results indicate that the nonlinearity improves the performance of the PageRank algorithm in terms of ranking effectiveness, as well as robustness against malicious manipulations. Although the nonlinearity analysis is based on the PageRank algorithm, it can be easily extended to other iterative ranking algorithms and similar improvements are expected.

Ranking scientific publications: the effect of nonlinearity.

Science.gov (United States)

Yao, Liyang; Wei, Tian; Zeng, An; Fan, Ying; Di, Zengru

2014-10-17

Ranking the significance of scientific publications is a long-standing challenge. The network-based analysis is a natural and common approach for evaluating the scientific credit of papers. Although the number of citations has been widely used as a metric to rank papers, recently some iterative processes such as the well-known PageRank algorithm have been applied to the citation networks to address this problem. In this paper, we introduce nonlinearity to the PageRank algorithm when aggregating resources from different nodes to further enhance the effect of important papers. The validation of our method is performed on the data of American Physical Society (APS) journals. The results indicate that the nonlinearity improves the performance of the PageRank algorithm in terms of ranking effectiveness, as well as robustness against malicious manipulations. Although the nonlinearity analysis is based on the PageRank algorithm, it can be easily extended to other iterative ranking algorithms and similar improvements are expected.

New second-order difference algorithm for image segmentation based on cellular neural networks (CNNs)

Science.gov (United States)

Meng, Shukai; Mo, Yu L.

2001-09-01

Image segmentation is one of the most important operations in many image analysis problems, which is the process that subdivides an image into its constituents and extracts those parts of interest. In this paper, we present a new second order difference gray-scale image segmentation algorithm based on cellular neural networks. A 3x3 CNN cloning template is applied, which can make smooth processing and has a good ability to deal with the conflict between the capability of noise resistance and the edge detection of complex shapes. We use second order difference operator to calculate the coefficients of the control template, which are not constant but rather depend on the input gray-scale values. It is similar to Contour Extraction CNN in construction, but there are some different in algorithm. The result of experiment shows that the second order difference CNN has a good capability in edge detection. It is better than Contour Extraction CNN in detail detection and more effective than the Laplacian of Gauss (LOG) algorithm.

A numerical approach for assessing effects of shear on equivalent permeability and nonlinear flow characteristics of 2-D fracture networks

Science.gov (United States)

Liu, Richeng; Li, Bo; Jiang, Yujing; Yu, Liyuan

2018-01-01

Hydro-mechanical properties of rock fractures are core issues for many geoscience and geo-engineering practices. Previous experimental and numerical studies have revealed that shear processes could greatly enhance the permeability of single rock fractures, yet the shear effects on hydraulic properties of fractured rock masses have received little attention. In most previous fracture network models, single fractures are typically presumed to be formed by parallel plates and flow is presumed to obey the cubic law. However, related studies have suggested that the parallel plate model cannot realistically represent the surface characters of natural rock fractures, and the relationship between flow rate and pressure drop will no longer be linear at sufficiently large Reynolds numbers. In the present study, a numerical approach was established to assess the effects of shear on the hydraulic properties of 2-D discrete fracture networks (DFNs) in both linear and nonlinear regimes. DFNs considering fracture surface roughness and variation of aperture in space were generated using an originally developed code DFNGEN. Numerical simulations by solving Navier-Stokes equations were performed to simulate the fluid flow through these DFNs. A fracture that cuts through each model was sheared and by varying the shear and normal displacements, effects of shear on equivalent permeability and nonlinear flow characteristics of DFNs were estimated. The results show that the critical condition of quantifying the transition from a linear flow regime to a nonlinear flow regime is: 10-4 〈 J hydraulic gradient. When the fluid flow is in a linear regime (i.e., J reduce the equivalent permeability significantly in the orientation perpendicular to the sheared fracture as much as 53.86% when J = 1, shear displacement Ds = 7 mm, and normal displacement Dn = 1 mm. By fitting the calculated results, the mathematical expression for δ2 is established to help choose proper governing equations when

Nonlinear predictive control for adaptive adjustments of deep brain stimulation parameters in basal ganglia-thalamic network.

Science.gov (United States)

Su, Fei; Wang, Jiang; Niu, Shuangxia; Li, Huiyan; Deng, Bin; Liu, Chen; Wei, Xile

2018-02-01

The efficacy of deep brain stimulation (DBS) for Parkinson's disease (PD) depends in part on the post-operative programming of stimulation parameters. Closed-loop stimulation is one method to realize the frequent adjustment of stimulation parameters. This paper introduced the nonlinear predictive control method into the online adjustment of DBS amplitude and frequency. This approach was tested in a computational model of basal ganglia-thalamic network. The autoregressive Volterra model was used to identify the process model based on physiological data. Simulation results illustrated the efficiency of closed-loop stimulation methods (amplitude adjustment and frequency adjustment) in improving the relay reliability of thalamic neurons compared with the PD state. Besides, compared with the 130Hz constant DBS the closed-loop stimulation methods can significantly reduce the energy consumption. Through the analysis of inter-spike-intervals (ISIs) distribution of basal ganglia neurons, the evoked network activity by the closed-loop frequency adjustment stimulation was closer to the normal state. Copyright © 2017 Elsevier Ltd. All rights reserved.

Numeral eddy current sensor modelling based on genetic neural network

International Nuclear Information System (INIS)

Yu Along

2008-01-01

This paper presents a method used to the numeral eddy current sensor modelling based on the genetic neural network to settle its nonlinear problem. The principle and algorithms of genetic neural network are introduced. In this method, the nonlinear model parameters of the numeral eddy current sensor are optimized by genetic neural network (GNN) according to measurement data. So the method remains both the global searching ability of genetic algorithm and the good local searching ability of neural network. The nonlinear model has the advantages of strong robustness, on-line modelling and high precision. The maximum nonlinearity error can be reduced to 0.037% by using GNN. However, the maximum nonlinearity error is 0.075% using the least square method

A spectral approach for the quantitative description of cardiac collagen network from nonlinear optical imaging.

Science.gov (United States)

Masè, Michela; Cristoforetti, Alessandro; Avogaro, Laura; Tessarolo, Francesco; Piccoli, Federico; Caola, Iole; Pederzolli, Carlo; Graffigna, Angelo; Ravelli, Flavia

2015-01-01

The assessment of collagen structure in cardiac pathology, such as atrial fibrillation (AF), is essential for a complete understanding of the disease. This paper introduces a novel methodology for the quantitative description of collagen network properties, based on the combination of nonlinear optical microscopy with a spectral approach of image processing and analysis. Second-harmonic generation (SHG) microscopy was applied to atrial tissue samples from cardiac surgery patients, providing label-free, selective visualization of the collagen structure. The spectral analysis framework, based on 2D-FFT, was applied to the SHG images, yielding a multiparametric description of collagen fiber orientation (angle and anisotropy indexes) and texture scale (dominant wavelength and peak dispersion indexes). The proof-of-concept application of the methodology showed the capability of our approach to detect and quantify differences in the structural properties of the collagen network in AF versus sinus rhythm patients. These results suggest the potential of our approach in the assessment of collagen properties in cardiac pathologies related to a fibrotic structural component.

Augmented Nonlinear Controller for Maximum Power-Point Tracking with Artificial Neural Network in Grid-Connected Photovoltaic Systems

Directory of Open Access Journals (Sweden)

Suliang Ma

2016-11-01

Full Text Available Photovoltaic (PV systems have non-linear characteristics that generate maximum power at one particular operating point. Environmental factors such as irradiance and temperature variations greatly affect the maximum power point (MPP. Diverse offline and online techniques have been introduced for tracking the MPP. Here, to track the MPP, an augmented-state feedback linearized (AFL non-linear controller combined with an artificial neural network (ANN is proposed. This approach linearizes the non-linear characteristics in PV systems and DC/DC converters, for tracking and optimizing the PV system operation. It also reduces the dependency of the designed controller on linearized models, to provide global stability. A complete model of the PV system is simulated. The existing maximum power-point tracking (MPPT and DC/DC boost-converter controller techniques are compared with the proposed ANN method. Two case studies, which simulate realistic circumstances, are presented to demonstrate the effectiveness and superiority of the proposed method. The AFL with ANN controller can provide good dynamic operation, faster convergence speed, and fewer operating-point oscillations around the MPP. It also tracks the global maxima under different conditions, especially irradiance-mutating situations, more effectively than the conventional methods. Detailed mathematical models and a control approach for a three-phase grid-connected intelligent hybrid system are proposed using MATLAB/Simulink.

Nonlinear Multiantenna Detection Methods

Directory of Open Access Journals (Sweden)

Chen Sheng

2004-01-01

Full Text Available A nonlinear detection technique designed for multiple-antenna assisted receivers employed in space-division multiple-access systems is investigated. We derive the optimal solution of the nonlinear spatial-processing assisted receiver for binary phase shift keying signalling, which we refer to as the Bayesian detector. It is shown that this optimal Bayesian receiver significantly outperforms the standard linear beamforming assisted receiver in terms of a reduced bit error rate, at the expense of an increased complexity, while the achievable system capacity is substantially enhanced with the advent of employing nonlinear detection. Specifically, when the spatial separation expressed in terms of the angle of arrival between the desired and interfering signals is below a certain threshold, a linear beamformer would fail to separate them, while a nonlinear detection assisted receiver is still capable of performing adequately. The adaptive implementation of the optimal Bayesian detector can be realized using a radial basis function network. Two techniques are presented for constructing block-data-based adaptive nonlinear multiple-antenna assisted receivers. One of them is based on the relevance vector machine invoked for classification, while the other on the orthogonal forward selection procedure combined with the Fisher ratio class-separability measure. A recursive sample-by-sample adaptation procedure is also proposed for training nonlinear detectors based on an amalgam of enhanced -means clustering techniques and the recursive least squares algorithm.

A comparison of linear and nonlinear statistical techniques in performance attribution.

Science.gov (United States)

Chan, N H; Genovese, C R

2001-01-01

Performance attribution is usually conducted under the linear framework of multifactor models. Although commonly used by practitioners in finance, linear multifactor models are known to be less than satisfactory in many situations. After a brief survey of nonlinear methods, nonlinear statistical techniques are applied to performance attribution of a portfolio constructed from a fixed universe of stocks using factors derived from some commonly used cross sectional linear multifactor models. By rebalancing this portfolio monthly, the cumulative returns for procedures based on standard linear multifactor model and three nonlinear techniques-model selection, additive models, and neural networks-are calculated and compared. It is found that the first two nonlinear techniques, especially in combination, outperform the standard linear model. The results in the neural-network case are inconclusive because of the great variety of possible models. Although these methods are more complicated and may require some tuning, toolboxes are developed and suggestions on calibration are proposed. This paper demonstrates the usefulness of modern nonlinear statistical techniques in performance attribution.

NONLINEAR ACCELERATOR LATTICES WITH ONE AND TWO ANALYTIC INVARIANTS

International Nuclear Information System (INIS)

Danilov, Viatcheslav V.

2010-01-01

Integrable systems appeared in physics long ago at the onset of classical dynamics with examples being Kepler s and other famous problems. Unfortunately, the majority of nonlinear problems turned out to be nonintegrable. In accelerator terms, any 2D nonlinear map produces a chaotic motion and a complex network of stable and unstable resonances with the unit probability. Nevertheless, in the proximity of an integrable system the full volume of such a chaotic network is small. Thus, the integrable nonlinear motion in accelerators has the potential to introduce a large betatron tune spread to suppress instabilities and to mitigate space charge effects with relatively small resonances and particle loss. To create such an accelerator lattice one has to find magnetic and electrtic field combinations leading to a stable integrable motion. This paper presents families of lattices with one invariant where bounded motion can be easily created in large volumes of the phase space. In addition, it presents 3 families of integrable nonlinear accelerator lattices, relizable with longitudinal-coordinate-dependent magnetic or electric fields with the stable nonlinear motion, which can be solved in terms of separable variables.

Optical stealth transmission based on super-continuum generation in highly nonlinear fiber over WDM network.

Science.gov (United States)

Zhu, Huatao; Wang, Rong; Pu, Tao; Fang, Tao; Xiang, Peng; Zheng, Jilin; Chen, Dalei

2015-06-01

In this Letter, the optical stealth transmission carried by super-continuum spectrum optical pulses generated in highly nonlinear fiber is proposed and experimentally demonstrated. In the proposed transmission scheme, super-continuum signals are reshaped in the spectral domain through a wavelength-selective switch and are temporally spread by a chromatic dispersion device to achieve the same noise-like characteristic as the noise in optical networks, so that in both the time domain and the spectral domain, the stealth signals are hidden in public channel. Our experimental results show that compared with existing schemes where stealth channels are carried by amplified spontaneous emission noise, super-continuum signal can increase the transmission performance and robustness.

Aggregated channels network for real-time pedestrian detection

Science.gov (United States)

Ghorban, Farzin; Marín, Javier; Su, Yu; Colombo, Alessandro; Kummert, Anton

2018-04-01

Convolutional neural networks (CNNs) have demonstrated their superiority in numerous computer vision tasks, yet their computational cost results prohibitive for many real-time applications such as pedestrian detection which is usually performed on low-consumption hardware. In order to alleviate this drawback, most strategies focus on using a two-stage cascade approach. Essentially, in the first stage a fast method generates a significant but reduced amount of high quality proposals that later, in the second stage, are evaluated by the CNN. In this work, we propose a novel detection pipeline that further benefits from the two-stage cascade strategy. More concretely, the enriched and subsequently compressed features used in the first stage are reused as the CNN input. As a consequence, a simpler network architecture, adapted for such small input sizes, allows to achieve real-time performance and obtain results close to the state-of-the-art while running significantly faster without the use of GPU. In particular, considering that the proposed pipeline runs in frame rate, the achieved performance is highly competitive. We furthermore demonstrate that the proposed pipeline on itself can serve as an effective proposal generator.

Performing particle image velocimetry using artificial neural networks: a proof-of-concept

Science.gov (United States)

Rabault, Jean; Kolaas, Jostein; Jensen, Atle

2017-12-01

Traditional programs based on feature engineering are underperforming on a steadily increasing number of tasks compared with artificial neural networks (ANNs), in particular for image analysis. Image analysis is widely used in fluid mechanics when performing particle image velocimetry (PIV) and particle tracking velocimetry (PTV), and therefore it is natural to test the ability of ANNs to perform such tasks. We report for the first time the use of convolutional neural networks (CNNs) and fully connected neural networks (FCNNs) for performing end-to-end PIV. Realistic synthetic images are used for training the networks and several synthetic test cases are used to assess the quality of each network’s predictions and compare them with state-of-the-art PIV software. In addition, we present tests on real-world data that prove ANNs can be used not only with synthetic images but also with more noisy, imperfect images obtained in a real experimental setup. While the ANNs we present have slightly higher root mean square error than state-of-the-art cross-correlation methods, they perform better near edges and allow for higher spatial resolution than such methods. In addition, it is likely that one could with further work develop ANNs which perform better that the proof-of-concept we offer.

Convolutional neural networks for event-related potential detection: impact of the architecture.

Science.gov (United States)

Cecotti, H

2017-07-01

The detection of brain responses at the single-trial level in the electroencephalogram (EEG) such as event-related potentials (ERPs) is a difficult problem that requires different processing steps to extract relevant discriminant features. While most of the signal and classification techniques for the detection of brain responses are based on linear algebra, different pattern recognition techniques such as convolutional neural network (CNN), as a type of deep learning technique, have shown some interests as they are able to process the signal after limited pre-processing. In this study, we propose to investigate the performance of CNNs in relation of their architecture and in relation to how they are evaluated: a single system for each subject, or a system for all the subjects. More particularly, we want to address the change of performance that can be observed between specifying a neural network to a subject, or by considering a neural network for a group of subjects, taking advantage of a larger number of trials from different subjects. The results support the conclusion that a convolutional neural network trained on different subjects can lead to an AUC above 0.9 by using an appropriate architecture using spatial filtering and shift invariant layers.

Observer-Based Controller Design for a Class of Nonlinear Networked Control Systems with Random Time-Delays Modeled by Markov Chains

Directory of Open Access Journals (Sweden)

Yanfeng Wang

2017-01-01

Full Text Available This paper investigates the observer-based controller design problem for a class of nonlinear networked control systems with random time-delays. The nonlinearity is assumed to satisfy a global Lipschitz condition and two dependent Markov chains are employed to describe the time-delay from sensor to controller (S-C delay and the time-delay from controller to actuator (C-A delay, respectively. The transition probabilities of S-C delay and C-A delay are both assumed to be partly inaccessible. Sufficient conditions on the stochastic stability for the closed-loop systems are obtained by constructing proper Lyapunov functional. The methods of calculating the controller and the observer gain matrix are also given. Two numerical examples are used to illustrate the effectiveness of the proposed method.

Nonlinearities in reservoir engineering: Enhancing quantum correlations

Science.gov (United States)

Hu, Xiangming; Hu, Qingping; Li, Lingchao; Huang, Chen; Rao, Shi

2017-12-01

There are two decisive factors for quantum correlations in reservoir engineering, but they are strongly reversely dependent on the atom-field nonlinearities. One is the squeezing parameter for the Bogoliubov modes-mediated collective interactions, while the other is the dissipative rates for the engineered collective dissipations. Exemplifying two-level atomic ensembles, we show that the moderate nonlinearities can compromise these two factors and thus enhance remarkably two-mode squeezing and entanglement of different spin atomic ensembles or different optical fields. This suggests that the moderate nonlinearities of the two-level systems are more advantageous for applications in quantum networks associated with reservoir engineering.

Combining deep residual neural network features with supervised machine learning algorithms to classify diverse food image datasets.

Science.gov (United States)

McAllister, Patrick; Zheng, Huiru; Bond, Raymond; Moorhead, Anne

2018-04-01

Obesity is increasing worldwide and can cause many chronic conditions such as type-2 diabetes, heart disease, sleep apnea, and some cancers. Monitoring dietary intake through food logging is a key method to maintain a healthy lifestyle to prevent and manage obesity. Computer vision methods have been applied to food logging to automate image classification for monitoring dietary intake. In this work we applied pretrained ResNet-152 and GoogleNet convolutional neural networks (CNNs), initially trained using ImageNet Large Scale Visual Recognition Challenge (ILSVRC) dataset with MatConvNet package, to extract features from food image datasets; Food 5K, Food-11, RawFooT-DB, and Food-101. Deep features were extracted from CNNs and used to train machine learning classifiers including artificial neural network (ANN), support vector machine (SVM), Random Forest, and Naive Bayes. Results show that using ResNet-152 deep features with SVM with RBF kernel can accurately detect food items with 99.4% accuracy using Food-5K validation food image dataset and 98.8% with Food-5K evaluation dataset using ANN, SVM-RBF, and Random Forest classifiers. Trained with ResNet-152 features, ANN can achieve 91.34%, 99.28% when applied to Food-11 and RawFooT-DB food image datasets respectively and SVM with RBF kernel can achieve 64.98% with Food-101 image dataset. From this research it is clear that using deep CNN features can be used efficiently for diverse food item image classification. The work presented in this research shows that pretrained ResNet-152 features provide sufficient generalisation power when applied to a range of food image classification tasks. Copyright © 2018 Elsevier Ltd. All rights reserved.

EDITORIAL: Nonlinear optical manipulation, patterning and control in nano- and micro-scale systems Nonlinear optical manipulation, patterning and control in nano- and micro-scale systems

Science.gov (United States)

Denz, Cornelia; Simoni, Francesco

2009-03-01

Nonlinearities are becoming more and more important for a variety of applications in nanosciences, bio-medical sciences, information processing and photonics. For applications at the crossings of these fields, especially microscopic and nanoscopic imaging and manipulation, nonlinearities play a key role. They may range from simple nonlinear parameter changes up to applications in manipulating, controlling and structuring material by light, or the manipulation of light by light itself. It is this area between basic nonlinear optics and photonic applications that includes `hot' topics such as ultra-resolution optical microscopy, micro- and nanomanipulation and -structuring, or nanophotonics. This special issue contains contributions in this field, many of them from the International Conference on Nonlinear Microscopy and Optical Control held in conjunction with a network meeting of the ESF COST action MP0604 `Optical Micromanipulation by Nonlinear Nanophotonics', 19-22 February 2008, Münster, Germany. Throughout this special issue, basic investigations of material structuring by nonlinear light--matter interaction, light-induced control of nanoparticles, and novel nonlinear material investigation techniques, are presented, covering the basic field of optical manipulation and control. These papers are followed by impressive developments of optical tweezers. Nowadays, optical phase contrast tweezers, twin and especially multiple beam traps, develop particle control in a new dimension: particles can be arranged, sorted and identified with high throughput. One of the most prominent forthcoming applications of optical tweezers is in the field of microfluidics. The action of light on fluids will open new horizons in microfluidic manipulation and control. The field of optical manipulation and control is a very broad field that has developed in an impressive way, in a short time, in Europe with the installation of the MP0604 network. Top researchers from 19 countries are

CNNs for sinusoidal signal recognition in hearing rehabilitation

Science.gov (United States)

Carnimeo, Leonarda; Giaquinto, Antonio

2003-04-01

In this paper, a contribution is given to provide a tool to the recognition of sinusoidal signals with a particular reference to the field of pediatric hearing rehabilitation. To this purpose, a synthesis technique previously developed by the authors' is used to design a Cellular Neural Network for an Associative Memory able to compare submitted discrete-time sinusoidal signals with memorized ones. A robustness analysis of the synthesized associative memory is also developed both for noisy inputs and for parameter variations. Simulation results are then reported to illustrate the performances of the designed network.

Comparison of the Gen Expression Programming, Nonlinear Time Series and Artificial Neural Network in Estimating the River Daily Flow (Case Study: The Karun River

Directory of Open Access Journals (Sweden)

R. Zamani

2015-06-01

Full Text Available Today, the daily flow forecasting of rivers is an important issue in hydrology and water resources and thus can be used the results of daily river flow modeling in water resources management, droughts and floods monitoring. In this study, due to the importance of this issue, using nonlinear time series models and artificial intelligence (Artificial Neural Network and Gen Expression Programming, the daily flow modeling has been at the time interval (1981-2012 in the Armand hydrometric station on the Karun River. Armand station upstream basin is one of the most basins in the North Karun basin and includes four sub basins (Vanak, Middle Karun, Beheshtabad and Kohrang.The results of this study shown that artificial intelligence models have superior than nonlinear time series in flow daily simulation in the Karun River. As well as, modeling and comparison of artificial intelligence models showed that the Gen Expression Programming have evaluation criteria better than artificial neural network.

An effective convolutional neural network model for Chinese sentiment analysis

Science.gov (United States)

Zhang, Yu; Chen, Mengdong; Liu, Lianzhong; Wang, Yadong

2017-06-01

Nowadays microblog is getting more and more popular. People are increasingly accustomed to expressing their opinions on Twitter, Facebook and Sina Weibo. Sentiment analysis of microblog has received significant attention, both in academia and in industry. So far, Chinese microblog exploration still needs lots of further work. In recent years CNN has also been used to deal with NLP tasks, and already achieved good results. However, these methods ignore the effective use of a large number of existing sentimental resources. For this purpose, we propose a Lexicon-based Sentiment Convolutional Neural Networks (LSCNN) model focus on Weibo's sentiment analysis, which combines two CNNs, trained individually base on sentiment features and word embedding, at the fully connected hidden layer. The experimental results show that our model outperforms the CNN model only with word embedding features on microblog sentiment analysis task.

Pattern Recognition of Momentary Mental Workload Based on Multi-Channel Electrophysiological Data and Ensemble Convolutional Neural Networks.

Science.gov (United States)

Zhang, Jianhua; Li, Sunan; Wang, Rubin

2017-01-01

In this paper, we deal with the Mental Workload (MWL) classification problem based on the measured physiological data. First we discussed the optimal depth (i.e., the number of hidden layers) and parameter optimization algorithms for the Convolutional Neural Networks (CNN). The base CNNs designed were tested according to five classification performance indices, namely Accuracy, Precision, F-measure, G-mean, and required training time. Then we developed an Ensemble Convolutional Neural Network (ECNN) to enhance the accuracy and robustness of the individual CNN model. For the ECNN design, three model aggregation approaches (weighted averaging, majority voting and stacking) were examined and a resampling strategy was used to enhance the diversity of individual CNN models. The results of MWL classification performance comparison indicated that the proposed ECNN framework can effectively improve MWL classification performance and is featured by entirely automatic feature extraction and MWL classification, when compared with traditional machine learning methods.

Global Detection of Live Virtual Machine Migration Based on Cellular Neural Networks

Directory of Open Access Journals (Sweden)

Kang Xie

2014-01-01

Full Text Available In order to meet the demands of operation monitoring of large scale, autoscaling, and heterogeneous virtual resources in the existing cloud computing, a new method of live virtual machine (VM migration detection algorithm based on the cellular neural networks (CNNs, is presented. Through analyzing the detection process, the parameter relationship of CNN is mapped as an optimization problem, in which improved particle swarm optimization algorithm based on bubble sort is used to solve the problem. Experimental results demonstrate that the proposed method can display the VM migration processing intuitively. Compared with the best fit heuristic algorithm, this approach reduces the processing time, and emerging evidence has indicated that this new approach is affordable to parallelism and analog very large scale integration (VLSI implementation allowing the VM migration detection to be performed better.

Convolutional neural network using generated data for SAR ATR with limited samples

Science.gov (United States)

Cong, Longjian; Gao, Lei; Zhang, Hui; Sun, Peng

2018-03-01

Being able to adapt all weather at all times, it has been a hot research topic that using Synthetic Aperture Radar(SAR) for remote sensing. Despite all the well-known advantages of SAR, it is hard to extract features because of its unique imaging methodology, and this challenge attracts the research interest of traditional Automatic Target Recognition(ATR) methods. With the development of deep learning technologies, convolutional neural networks(CNNs) give us another way out to detect and recognize targets, when a huge number of samples are available, but this premise is often not hold, when it comes to monitoring a specific type of ships. In this paper, we propose a method to enhance the performance of Faster R-CNN with limited samples to detect and recognize ships in SAR images.

Global detection of live virtual machine migration based on cellular neural networks.

Science.gov (United States)

Xie, Kang; Yang, Yixian; Zhang, Ling; Jing, Maohua; Xin, Yang; Li, Zhongxian

2014-01-01

In order to meet the demands of operation monitoring of large scale, autoscaling, and heterogeneous virtual resources in the existing cloud computing, a new method of live virtual machine (VM) migration detection algorithm based on the cellular neural networks (CNNs), is presented. Through analyzing the detection process, the parameter relationship of CNN is mapped as an optimization problem, in which improved particle swarm optimization algorithm based on bubble sort is used to solve the problem. Experimental results demonstrate that the proposed method can display the VM migration processing intuitively. Compared with the best fit heuristic algorithm, this approach reduces the processing time, and emerging evidence has indicated that this new approach is affordable to parallelism and analog very large scale integration (VLSI) implementation allowing the VM migration detection to be performed better.

A network dynamics approach to chemical reaction networks

NARCIS (Netherlands)

van der Schaft, Abraham; Rao, S.; Jayawardhana, B.

2016-01-01

A treatment of chemical reaction network theory is given from the perspective of nonlinear network dynamics, in particular of consensus dynamics. By starting from the complex-balanced assumption the reaction dynamics governed by mass action kinetics can be rewritten into a form which allows for a

Synchronization of networks

Indian Academy of Sciences (India)

We study the synchronization of coupled dynamical systems on networks. The dynamics is governed by a local nonlinear oscillator for each node of the network and interactions connecting different nodes via the links of the network. We consider existence and stability conditions for both single- and multi-cluster ...

Hyperspectral Image Enhancement and Mixture Deep-Learning Classification of Corneal Epithelium Injuries.

Science.gov (United States)

Noor, Siti Salwa Md; Michael, Kaleena; Marshall, Stephen; Ren, Jinchang

2017-11-16

In our preliminary study, the reflectance signatures obtained from hyperspectral imaging (HSI) of normal and abnormal corneal epithelium tissues of porcine show similar morphology with subtle differences. Here we present image enhancement algorithms that can be used to improve the interpretability of data into clinically relevant information to facilitate diagnostics. A total of 25 corneal epithelium images without the application of eye staining were used. Three image feature extraction approaches were applied for image classification: (i) image feature classification from histogram using a support vector machine with a Gaussian radial basis function (SVM-GRBF); (ii) physical image feature classification using deep-learning Convolutional Neural Networks (CNNs) only; and (iii) the combined classification of CNNs and SVM-Linear. The performance results indicate that our chosen image features from the histogram and length-scale parameter were able to classify with up to 100% accuracy; particularly, at CNNs and CNNs-SVM, by employing 80% of the data sample for training and 20% for testing. Thus, in the assessment of corneal epithelium injuries, HSI has high potential as a method that could surpass current technologies regarding speed, objectivity, and reliability.

Noise and non-linearities in high-throughput data

International Nuclear Information System (INIS)

Nguyen, Viet-Anh; Lió, Pietro; Koukolíková-Nicola, Zdena; Bagnoli, Franco

2009-01-01

High-throughput data analyses are becoming common in biology, communications, economics and sociology. The vast amounts of data are usually represented in the form of matrices and can be considered as knowledge networks. Spectra-based approaches have proved useful in extracting hidden information within such networks and for estimating missing data, but these methods are based essentially on linear assumptions. The physical models of matching, when applicable, often suggest non-linear mechanisms, that may sometimes be identified as noise. The use of non-linear models in data analysis, however, may require the introduction of many parameters, which lowers the statistical weight of the model. According to the quality of data, a simpler linear analysis may be more convenient than more complex approaches. In this paper, we show how a simple non-parametric Bayesian model may be used to explore the role of non-linearities and noise in synthetic and experimental data sets

Applications of Nonlinear Dynamics Model and Design of Complex Systems

CERN Document Server

In, Visarath; Palacios, Antonio

2009-01-01

This edited book is aimed at interdisciplinary, device-oriented, applications of nonlinear science theory and methods in complex systems. In particular, applications directed to nonlinear phenomena with space and time characteristics. Examples include: complex networks of magnetic sensor systems, coupled nano-mechanical oscillators, nano-detectors, microscale devices, stochastic resonance in multi-dimensional chaotic systems, biosensors, and stochastic signal quantization. "applications of nonlinear dynamics: model and design of complex systems" brings together the work of scientists and engineers that are applying ideas and methods from nonlinear dynamics to design and fabricate complex systems.

XXIII International Conference on Nonlinear Dynamics of Electronic Systems

CERN Document Server

Stoop, Ruedi; Stramaglia, Sebastiano

2017-01-01

This book collects contributions to the XXIII international conference “Nonlinear dynamics of electronic systems”. Topics range from non-linearity in electronic circuits to synchronisation effects in complex networks to biological systems, neural dynamics and the complex organisation of the brain. Resting on a solid mathematical basis, these investigations address highly interdisciplinary problems in physics, engineering, biology and biochemistry.

Reconstruction of Micropattern Detector Signals using Convolutional Neural Networks

Science.gov (United States)

Flekova, L.; Schott, M.

2017-10-01

Micropattern gaseous detector (MPGD) technologies, such as GEMs or MicroMegas, are particularly suitable for precision tracking and triggering in high rate environments. Given their relatively low production costs, MPGDs are an exemplary candidate for the next generation of particle detectors. Having acknowledged these advantages, both the ATLAS and CMS collaborations at the LHC are exploiting these new technologies for their detector upgrade programs in the coming years. When MPGDs are utilized for triggering purposes, the measured signals need to be precisely reconstructed within less than 200 ns, which can be achieved by the usage of FPGAs. In this work, we present a novel approach to identify reconstructed signals, their timing and the corresponding spatial position on the detector. In particular, we study the effect of noise and dead readout strips on the reconstruction performance. Our approach leverages the potential of convolutional neural network (CNNs), which have recently manifested an outstanding performance in a range of modeling tasks. The proposed neural network architecture of our CNN is designed simply enough, so that it can be modeled directly by an FPGA and thus provide precise information on reconstructed signals already in trigger level.

Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification

Directory of Open Access Journals (Sweden)

Srdjan Sladojevic

2016-01-01

Full Text Available The latest generation of convolutional neural networks (CNNs has achieved impressive results in the field of image classification. This paper is concerned with a new approach to the development of plant disease recognition model, based on leaf image classification, by the use of deep convolutional networks. Novel way of training and the methodology used facilitate a quick and easy system implementation in practice. The developed model is able to recognize 13 different types of plant diseases out of healthy leaves, with the ability to distinguish plant leaves from their surroundings. According to our knowledge, this method for plant disease recognition has been proposed for the first time. All essential steps required for implementing this disease recognition model are fully described throughout the paper, starting from gathering images in order to create a database, assessed by agricultural experts. Caffe, a deep learning framework developed by Berkley Vision and Learning Centre, was used to perform the deep CNN training. The experimental results on the developed model achieved precision between 91% and 98%, for separate class tests, on average 96.3%.

Generalized Nonlinear Yule Models

Science.gov (United States)

Lansky, Petr; Polito, Federico; Sacerdote, Laura

2016-11-01

With the aim of considering models related to random graphs growth exhibiting persistent memory, we propose a fractional nonlinear modification of the classical Yule model often studied in the context of macroevolution. Here the model is analyzed and interpreted in the framework of the development of networks such as the World Wide Web. Nonlinearity is introduced by replacing the linear birth process governing the growth of the in-links of each specific webpage with a fractional nonlinear birth process with completely general birth rates. Among the main results we derive the explicit distribution of the number of in-links of a webpage chosen uniformly at random recognizing the contribution to the asymptotics and the finite time correction. The mean value of the latter distribution is also calculated explicitly in the most general case. Furthermore, in order to show the usefulness of our results, we particularize them in the case of specific birth rates giving rise to a saturating behaviour, a property that is often observed in nature. The further specialization to the non-fractional case allows us to extend the Yule model accounting for a nonlinear growth.

Real-time neural network-based self-tuning control of a nonlinear electro-hydraulic servomotor

Energy Technology Data Exchange (ETDEWEB)

Canelon, J.I.; Ortega, A.G. [Univ. del Zulia, Maracaibo, Zulia (Venezuela, Bolivarian Republic of). School of Electrical Engineering; Shieh, L.S. [Houston Univ., Houston, TX (United States). Dept. of Electrical and Computer Engineering; Bastidas, J.I. [Univ. del Zulia, Maracaibo, Zulia (Venezuela, Bolivarian Republic of). School of Mechanical Engineering; Zhang, Y.; Akujuobi, C.M. [Prairie View A and M Univ., Prairie View, TX (United States). Center of Excellence for Communication Systems Technology Research and Dept. of Engineering Technology

2010-08-13

For high power applications, hydraulic actuators offer many advantages over electromagnetic actuators, including higher torque/mass ratios; smaller control gains; excellent torque capability; filtered high frequency noise; better heat transfer characteristics; smaller size; higher speed of response of the servomechanism; cheaper hardware; and higher reliability. Therefore, any application that requires a large force applied smoothly by an actuator is a candidate for hydraulic power. Examples of such applications include vehicle steering and braking systems; roll mills; drilling rigs; heavy duty crane and presses; and industrial robots and actuators for aircraft control surfaces such as ailerons and flaps. It is extremely important to create effective control strategies for hydraulic systems. This paper outlined the real-time implementation of a neural network-based approach, for self-tuning control of the angular position of a nonlinear electro-hydraulic servomotor. Using an online training algorithm, a neural network autoregressive moving-average model with exogenous input (ARMAX) model of the system was identified and continuously updated and an optimal linear ARMAX model was determined. The paper briefly depicted the neural network-based self-tuning control approach and a description of the experimental equipment (hardware and software) was presented including the implementation details. The experimental results were discussed and conclusions were summarized. It was found that the approach proved to be very effective in the control of this fast dynamics system, outperforming a fine tuned PI controller. Therefore, although the self-tuning approach was computationally demanding, it was feasible for real-time implementation. 22 refs., 6 figs.

Brain tumor segmentation with Deep Neural Networks.

Science.gov (United States)

Havaei, Mohammad; Davy, Axel; Warde-Farley, David; Biard, Antoine; Courville, Aaron; Bengio, Yoshua; Pal, Chris; Jodoin, Pierre-Marc; Larochelle, Hugo

2017-01-01

In this paper, we present a fully automatic brain tumor segmentation method based on Deep Neural Networks (DNNs). The proposed networks are tailored to glioblastomas (both low and high grade) pictured in MR images. By their very nature, these tumors can appear anywhere in the brain and have almost any kind of shape, size, and contrast. These reasons motivate our exploration of a machine learning solution that exploits a flexible, high capacity DNN while being extremely efficient. Here, we give a description of different model choices that we've found to be necessary for obtaining competitive performance. We explore in particular different architectures based on Convolutional Neural Networks (CNN), i.e. DNNs specifically adapted to image data. We present a novel CNN architecture which differs from those traditionally used in computer vision. Our CNN exploits both local features as well as more global contextual features simultaneously. Also, different from most traditional uses of CNNs, our networks use a final layer that is a convolutional implementation of a fully connected layer which allows a 40 fold speed up. We also describe a 2-phase training procedure that allows us to tackle difficulties related to the imbalance of tumor labels. Finally, we explore a cascade architecture in which the output of a basic CNN is treated as an additional source of information for a subsequent CNN. Results reported on the 2013 BRATS test data-set reveal that our architecture improves over the currently published state-of-the-art while being over 30 times faster. Copyright © 2016 Elsevier B.V. All rights reserved.

modeling and optimization of an electric power distribution network

African Journals Online (AJOL)

user

EDNEPP) was solved by a mixed binary integer programming (MBIP) formulation of the network, where the steady-state operation of the network was modelled with non-linear mathematical expressions. The non-linear terms are linearized, using ...

A single charge in the actin binding domain of fascin can independently tune the linear and non-linear response of an actin bundle network.

Science.gov (United States)

Maier, M; Müller, K W; Heussinger, C; Köhler, S; Wall, W A; Bausch, A R; Lieleg, O

2015-05-01

Actin binding proteins (ABPs) not only set the structure of actin filament assemblies but also mediate the frequency-dependent viscoelastic moduli of cross-linked and bundled actin networks. Point mutations in the actin binding domain of those ABPs can tune the association and dissociation dynamics of the actin/ABP bond and thus modulate the network mechanics both in the linear and non-linear response regime. We here demonstrate how the exchange of a single charged amino acid in the actin binding domain of the ABP fascin triggers such a modulation of the network rheology. Whereas the overall structure of the bundle networks is conserved, the transition point from strain-hardening to strain-weakening sensitively depends on the cross-linker off-rate and the applied shear rate. Our experimental results are consistent both with numerical simulations of a cross-linked bundle network and a theoretical description of the bundle network mechanics which is based on non-affine bending deformations and force-dependent cross-link dynamics.

Empirical modeling of nuclear power plants using neural networks

International Nuclear Information System (INIS)

Parlos, A.G.; Atiya, A.; Chong, K.T.

1991-01-01

A summary of a procedure for nonlinear identification of process dynamics encountered in nuclear power plant components is presented in this paper using artificial neural systems. A hybrid feedforward/feedback neural network, namely, a recurrent multilayer perceptron, is used as the nonlinear structure for system identification. In the overall identification process, the feedforward portion of the network architecture provides its well-known interpolation property, while through recurrency and cross-talk, the local information feedback enables representation of time-dependent system nonlinearities. The standard backpropagation learning algorithm is modified and is used to train the proposed hybrid network in a supervised manner. The performance of recurrent multilayer perceptron networks in identifying process dynamics is investigated via the case study of a U-tube steam generator. The nonlinear response of a representative steam generator is predicted using a neural network and is compared to the response obtained from a sophisticated physical model during both high- and low-power operation. The transient responses compare well, though further research is warranted for training and testing of recurrent neural networks during more severe operational transients and accident scenarios

Calibration of Mine Ventilation Network Models Using the Non-Linear Optimization Algorithm

Directory of Open Access Journals (Sweden)

Guang Xu

2017-12-01

Full Text Available Effective ventilation planning is vital to underground mining. To ensure stable operation of the ventilation system and to avoid airflow disorder, mine ventilation network (MVN models have been widely used in simulating and optimizing the mine ventilation system. However, one of the challenges for MVN model simulation is that the simulated airflow distribution results do not match the measured data. To solve this problem, a simple and effective calibration method is proposed based on the non-linear optimization algorithm. The calibrated model not only makes simulated airflow distribution results in accordance with the on-site measured data, but also controls the errors of other parameters within a minimum range. The proposed method was then applied to calibrate an MVN model in a real case, which is built based on ventilation survey results and Ventsim software. Finally, airflow simulation experiments are carried out respectively using data before and after calibration, whose results were compared and analyzed. This showed that the simulated airflows in the calibrated model agreed much better to the ventilation survey data, which verifies the effectiveness of calibrating method.

Distributed control design for nonlinear output agreement in convergent systems

NARCIS (Netherlands)

Weitenberg, Erik; De Persis, Claudio

2015-01-01

This work studies the problem of output agreement in homogeneous networks of nonlinear dynamical systems under time-varying disturbances using controllers placed at the nodes of the networks. For the class of contractive systems, necessary and sufficient conditions for output agreement are derived,

First Steps Toward Incorporating Image Based Diagnostics Into Particle Accelerator Control Systems Using Convolutional Neural Networks

Energy Technology Data Exchange (ETDEWEB)

Edelen, A. L.; Biedron, S. G.; Milton, S. V.; Edelen, J. P.

2016-12-16

At present, a variety of image-based diagnostics are used in particle accelerator systems. Often times, these are viewed by a human operator who then makes appropriate adjustments to the machine. Given recent advances in using convolutional neural networks (CNNs) for image processing, it should be possible to use image diagnostics directly in control routines (NN-based or otherwise). This is especially appealing for non-intercepting diagnostics that could run continuously during beam operation. Here, we show results of a first step toward implementing such a controller: our trained CNN can predict multiple simulated downstream beam parameters at the Fermilab Accelerator Science and Technology (FAST) facility's low energy beamline using simulated virtual cathode laser images, gun phases, and solenoid strengths.

Limitations and tradeoffs in synchronization of large-scale networks with uncertain links

Science.gov (United States)

Diwadkar, Amit; Vaidya, Umesh

2016-01-01

The synchronization of nonlinear systems connected over large-scale networks has gained popularity in a variety of applications, such as power grids, sensor networks, and biology. Stochastic uncertainty in the interconnections is a ubiquitous phenomenon observed in these physical and biological networks. We provide a size-independent network sufficient condition for the synchronization of scalar nonlinear systems with stochastic linear interactions over large-scale networks. This sufficient condition, expressed in terms of nonlinear dynamics, the Laplacian eigenvalues of the nominal interconnections, and the variance and location of the stochastic uncertainty, allows us to define a synchronization margin. We provide an analytical characterization of important trade-offs between the internal nonlinear dynamics, network topology, and uncertainty in synchronization. For nearest neighbour networks, the existence of an optimal number of neighbours with a maximum synchronization margin is demonstrated. An analytical formula for the optimal gain that produces the maximum synchronization margin allows us to compare the synchronization properties of various complex network topologies. PMID:27067994

CNNdel: Calling Structural Variations on Low Coverage Data Based on Convolutional Neural Networks

Directory of Open Access Journals (Sweden)

Jing Wang

2017-01-01

Full Text Available Many structural variations (SVs detection methods have been proposed due to the popularization of next-generation sequencing (NGS. These SV calling methods use different SV-property-dependent features; however, they all suffer from poor accuracy when running on low coverage sequences. The union of results from these tools achieves fairly high sensitivity but still produces low accuracy on low coverage sequence data. That is, these methods contain many false positives. In this paper, we present CNNdel, an approach for calling deletions from paired-end reads. CNNdel gathers SV candidates reported by multiple tools and then extracts features from aligned BAM files at the positions of candidates. With labeled feature-expressed candidates as a training set, CNNdel trains convolutional neural networks (CNNs to distinguish true unlabeled candidates from false ones. Results show that CNNdel works well with NGS reads from 26 low coverage genomes of the 1000 Genomes Project. The paper demonstrates that convolutional neural networks can automatically assign the priority of SV features and reduce the false positives efficaciously.

Neural network based online simultaneous policy update algorithm for solving the HJI equation in nonlinear H∞ control.

Science.gov (United States)

Wu, Huai-Ning; Luo, Biao

2012-12-01

It is well known that the nonlinear H∞ state feedback control problem relies on the solution of the Hamilton-Jacobi-Isaacs (HJI) equation, which is a nonlinear partial differential equation that has proven to be impossible to solve analytically. In this paper, a neural network (NN)-based online simultaneous policy update algorithm (SPUA) is developed to solve the HJI equation, in which knowledge of internal system dynamics is not required. First, we propose an online SPUA which can be viewed as a reinforcement learning technique for two players to learn their optimal actions in an unknown environment. The proposed online SPUA updates control and disturbance policies simultaneously; thus, only one iterative loop is needed. Second, the convergence of the online SPUA is established by proving that it is mathematically equivalent to Newton's method for finding a fixed point in a Banach space. Third, we develop an actor-critic structure for the implementation of the online SPUA, in which only one critic NN is needed for approximating the cost function, and a least-square method is given for estimating the NN weight parameters. Finally, simulation studies are provided to demonstrate the effectiveness of the proposed algorithm.

Nonlinear mechanical response of the extracellular matrix: learning from articular cartilage

Science.gov (United States)

Kearns, Sarah; Das, Moumita

2015-03-01

We study the mechanical structure-function relations in the extracellular matrix (ECM) with focus on nonlinear shear and compression response. As a model system, our study focuses on the ECM in articular cartilage tissue which has two major mechanobiological components: a network of the biopolymer collagen that acts as a stiff, reinforcing matrix, and a flexible aggrecan network that facilitates deformability. We model this system as a double network hydrogel made of interpenetrating networks of stiff and flexible biopolymers respectively. We study the linear and nonlinear mechanical response of the model ECM to shear and compression forces using a combination of rigidity percolation theory and energy minimization approaches. Our results may provide useful insights into the design principles of the ECM as well as biomimetic hydrogels that are mechanically robust and can, at the same time, easily adapt to cues in their surroundings.

Forecasting with nonlinear time series models

DEFF Research Database (Denmark)

Kock, Anders Bredahl; Teräsvirta, Timo

In this paper, nonlinear models are restricted to mean nonlinear parametric models. Several such models popular in time series econo- metrics are presented and some of their properties discussed. This in- cludes two models based on universal approximators: the Kolmogorov- Gabor polynomial model...... applied to economic fore- casting problems, is briefly highlighted. A number of large published studies comparing macroeconomic forecasts obtained using different time series models are discussed, and the paper also contains a small simulation study comparing recursive and direct forecasts in a partic...... and two versions of a simple artificial neural network model. Techniques for generating multi-period forecasts from nonlinear models recursively are considered, and the direct (non-recursive) method for this purpose is mentioned as well. Forecasting with com- plex dynamic systems, albeit less frequently...

Spatio-temporal analysis of brain electrical activity in epilepsy based on cellular nonlinear networks

Science.gov (United States)

Gollas, Frank; Tetzlaff, Ronald

2009-05-01

Epilepsy is the most common chronic disorder of the nervous system. Generally, epileptic seizures appear without foregoing sign or warning. The problem of detecting a possible pre-seizure state in epilepsy from EEG signals has been addressed by many authors over the past decades. Different approaches of time series analysis of brain electrical activity already are providing valuable insights into the underlying complex dynamics. But the main goal the identification of an impending epileptic seizure with a sufficient specificity and reliability, has not been achieved up to now. An algorithm for a reliable, automated prediction of epileptic seizures would enable the realization of implantable seizure warning devices, which could provide valuable information to the patient and time/event specific drug delivery or possibly a direct electrical nerve stimulation. Cellular Nonlinear Networks (CNN) are promising candidates for future seizure warning devices. CNN are characterized by local couplings of comparatively simple dynamical systems. With this property these networks are well suited to be realized as highly parallel, analog computer chips. Today available CNN hardware realizations exhibit a processing speed in the range of TeraOps combined with low power consumption. In this contribution new algorithms based on the spatio-temporal dynamics of CNN are considered in order to analyze intracranial EEG signals and thus taking into account mutual dependencies between neighboring regions of the brain. In an identification procedure Reaction-Diffusion CNN (RD-CNN) are determined for short segments of brain electrical activity, by means of a supervised parameter optimization. RD-CNN are deduced from Reaction-Diffusion Systems, which usually are applied to investigate complex phenomena like nonlinear wave propagation or pattern formation. The Local Activity Theory provides a necessary condition for emergent behavior in RD-CNN. In comparison linear spatio

Data based identification and prediction of nonlinear and complex dynamical systems

Science.gov (United States)

Wang, Wen-Xu; Lai, Ying-Cheng; Grebogi, Celso

2016-07-01

systems theories with tools from statistical physics, optimization, engineering control, applied mathematics, and scientific computing enables the development of a number of paradigms to address the problem of nonlinear and complex systems reconstruction. In this Review, we describe the recent advances in this forefront and rapidly evolving field, with a focus on compressive sensing based methods. In particular, compressive sensing is a paradigm developed in recent years in applied mathematics, electrical engineering, and nonlinear physics to reconstruct sparse signals using only limited data. It has broad applications ranging from image compression/reconstruction to the analysis of large-scale sensor networks, and it has become a powerful technique to obtain high-fidelity signals for applications where sufficient observations are not available. We will describe in detail how compressive sensing can be exploited to address a diverse array of problems in data based reconstruction of nonlinear and complex networked systems. The problems include identification of chaotic systems and prediction of catastrophic bifurcations, forecasting future attractors of time-varying nonlinear systems, reconstruction of complex networks with oscillatory and evolutionary game dynamics, detection of hidden nodes, identification of chaotic elements in neuronal networks, reconstruction of complex geospatial networks and nodal positioning, and reconstruction of complex spreading networks with binary data.. A number of alternative methods, such as those based on system response to external driving, synchronization, and noise-induced dynamical correlation, will also be discussed. Due to the high relevance of network reconstruction to biological sciences, a special section is devoted to a brief survey of the current methods to infer biological networks. Finally, a number of open problems including control and controllability of complex nonlinear dynamical networks are discussed. The methods

Data based identification and prediction of nonlinear and complex dynamical systems

Energy Technology Data Exchange (ETDEWEB)

Wang, Wen-Xu [School of Systems Science, Beijing Normal University, Beijing, 100875 (China); Business School, University of Shanghai for Science and Technology, Shanghai 200093 (China); Lai, Ying-Cheng, E-mail: Ying-Cheng.Lai@asu.edu [School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287 (United States); Department of Physics, Arizona State University, Tempe, AZ 85287 (United States); Institute for Complex Systems and Mathematical Biology, King’s College, University of Aberdeen, Aberdeen AB24 3UE (United Kingdom); Grebogi, Celso [Institute for Complex Systems and Mathematical Biology, King’s College, University of Aberdeen, Aberdeen AB24 3UE (United Kingdom)

2016-07-12

dynamical systems theories with tools from statistical physics, optimization, engineering control, applied mathematics, and scientific computing enables the development of a number of paradigms to address the problem of nonlinear and complex systems reconstruction. In this Review, we describe the recent advances in this forefront and rapidly evolving field, with a focus on compressive sensing based methods. In particular, compressive sensing is a paradigm developed in recent years in applied mathematics, electrical engineering, and nonlinear physics to reconstruct sparse signals using only limited data. It has broad applications ranging from image compression/reconstruction to the analysis of large-scale sensor networks, and it has become a powerful technique to obtain high-fidelity signals for applications where sufficient observations are not available. We will describe in detail how compressive sensing can be exploited to address a diverse array of problems in data based reconstruction of nonlinear and complex networked systems. The problems include identification of chaotic systems and prediction of catastrophic bifurcations, forecasting future attractors of time-varying nonlinear systems, reconstruction of complex networks with oscillatory and evolutionary game dynamics, detection of hidden nodes, identification of chaotic elements in neuronal networks, reconstruction of complex geospatial networks and nodal positioning, and reconstruction of complex spreading networks with binary data.. A number of alternative methods, such as those based on system response to external driving, synchronization, and noise-induced dynamical correlation, will also be discussed. Due to the high relevance of network reconstruction to biological sciences, a special section is devoted to a brief survey of the current methods to infer biological networks. Finally, a number of open problems including control and controllability of complex nonlinear dynamical networks are discussed. The methods

Data based identification and prediction of nonlinear and complex dynamical systems

International Nuclear Information System (INIS)

Wang, Wen-Xu; Lai, Ying-Cheng; Grebogi, Celso

2016-01-01

dynamical systems theories with tools from statistical physics, optimization, engineering control, applied mathematics, and scientific computing enables the development of a number of paradigms to address the problem of nonlinear and complex systems reconstruction. In this Review, we describe the recent advances in this forefront and rapidly evolving field, with a focus on compressive sensing based methods. In particular, compressive sensing is a paradigm developed in recent years in applied mathematics, electrical engineering, and nonlinear physics to reconstruct sparse signals using only limited data. It has broad applications ranging from image compression/reconstruction to the analysis of large-scale sensor networks, and it has become a powerful technique to obtain high-fidelity signals for applications where sufficient observations are not available. We will describe in detail how compressive sensing can be exploited to address a diverse array of problems in data based reconstruction of nonlinear and complex networked systems. The problems include identification of chaotic systems and prediction of catastrophic bifurcations, forecasting future attractors of time-varying nonlinear systems, reconstruction of complex networks with oscillatory and evolutionary game dynamics, detection of hidden nodes, identification of chaotic elements in neuronal networks, reconstruction of complex geospatial networks and nodal positioning, and reconstruction of complex spreading networks with binary data.. A number of alternative methods, such as those based on system response to external driving, synchronization, and noise-induced dynamical correlation, will also be discussed. Due to the high relevance of network reconstruction to biological sciences, a special section is devoted to a brief survey of the current methods to infer biological networks. Finally, a number of open problems including control and controllability of complex nonlinear dynamical networks are discussed. The methods

NNSYSID and NNCTRL Tools for system identification and control with neural networks

DEFF Research Database (Denmark)

Nørgaard, Magnus; Ravn, Ole; Poulsen, Niels Kjølstad

2001-01-01

choose among several designs such as direct inverse control, internal model control, nonlinear feedforward, feedback linearisation, optimal control, gain scheduling based on instantaneous linearisation of neural network models and nonlinear model predictive control. This article gives an overview......Two toolsets for use with MATLAB have been developed: the neural network based system identification toolbox (NNSYSID) and the neural network based control system design toolkit (NNCTRL). The NNSYSID toolbox has been designed to assist identification of nonlinear dynamic systems. It contains...... a number of nonlinear model structures based on neural networks, effective training algorithms and tools for model validation and model structure selection. The NNCTRL toolkit is an add-on to NNSYSID and provides tools for design and simulation of control systems based on neural networks. The user can...

NNSYSID and NNCTRL Tools for system identification and control with neural networks

DEFF Research Database (Denmark)

Nørgaard, Magnus; Ravn, Ole; Poulsen, Niels Kjølstad

2001-01-01

a number of nonlinear model structures based on neural networks, effective training algorithms and tools for model validation and model structure selection. The NNCTRL toolkit is an add-on to NNSYSID and provides tools for design and simulation of control systems based on neural networks. The user can...... choose among several designs such as direct inverse control, internal model control, nonlinear feedforward, feedback linearisation, optimal control, gain scheduling based on instantaneous linearisation of neural network models and nonlinear model predictive control. This article gives an overview......Two toolsets for use with MATLAB have been developed: the neural network based system identification toolbox (NNSYSID) and the neural network based control system design toolkit (NNCTRL). The NNSYSID toolbox has been designed to assist identification of nonlinear dynamic systems. It contains...

Constructing fine-granularity functional brain network atlases via deep convolutional autoencoder.

Science.gov (United States)

Zhao, Yu; Dong, Qinglin; Chen, Hanbo; Iraji, Armin; Li, Yujie; Makkie, Milad; Kou, Zhifeng; Liu, Tianming

2017-12-01

State-of-the-art functional brain network reconstruction methods such as independent component analysis (ICA) or sparse coding of whole-brain fMRI data can effectively infer many thousands of volumetric brain network maps from a large number of human brains. However, due to the variability of individual brain networks and the large scale of such networks needed for statistically meaningful group-level analysis, it is still a challenging and open problem to derive group-wise common networks as network atlases. Inspired by the superior spatial pattern description ability of the deep convolutional neural networks (CNNs), a novel deep 3D convolutional autoencoder (CAE) network is designed here to extract spatial brain network features effectively, based on which an Apache Spark enabled computational framework is developed for fast clustering of larger number of network maps into fine-granularity atlases. To evaluate this framework, 10 resting state networks (RSNs) were manually labeled from the sparsely decomposed networks of Human Connectome Project (HCP) fMRI data and 5275 network training samples were obtained, in total. Then the deep CAE models are trained by these functional networks' spatial maps, and the learned features are used to refine the original 10 RSNs into 17 network atlases that possess fine-granularity functional network patterns. Interestingly, it turned out that some manually mislabeled outliers in training networks can be corrected by the deep CAE derived features. More importantly, fine granularities of networks can be identified and they reveal unique network patterns specific to different brain task states. By further applying this method to a dataset of mild traumatic brain injury study, it shows that the technique can effectively identify abnormal small networks in brain injury patients in comparison with controls. In general, our work presents a promising deep learning and big data analysis solution for modeling functional connectomes, with

Multiscale deep features learning for land-use scene recognition

Science.gov (United States)

Yuan, Baohua; Li, Shijin; Li, Ning

2018-01-01

The features extracted from deep convolutional neural networks (CNNs) have shown their promise as generic descriptors for land-use scene recognition. However, most of the work directly adopts the deep features for the classification of remote sensing images, and does not encode the deep features for improving their discriminative power, which can affect the performance of deep feature representations. To address this issue, we propose an effective framework, LASC-CNN, obtained by locality-constrained affine subspace coding (LASC) pooling of a CNN filter bank. LASC-CNN obtains more discriminative deep features than directly extracted from CNNs. Furthermore, LASC-CNN builds on the top convolutional layers of CNNs, which can incorporate multiscale information and regions of arbitrary resolution and sizes. Our experiments have been conducted using two widely used remote sensing image databases, and the results show that the proposed method significantly improves the performance when compared to other state-of-the-art methods.

Modified Convolutional Neural Network Based on Dropout and the Stochastic Gradient Descent Optimizer

Directory of Open Access Journals (Sweden)

Jing Yang

2018-03-01

Full Text Available This study proposes a modified convolutional neural network (CNN algorithm that is based on dropout and the stochastic gradient descent (SGD optimizer (MCNN-DS, after analyzing the problems of CNNs in extracting the convolution features, to improve the feature recognition rate and reduce the time-cost of CNNs. The MCNN-DS has a quadratic CNN structure and adopts the rectified linear unit as the activation function to avoid the gradient problem and accelerate convergence. To address the overfitting problem, the algorithm uses an SGD optimizer, which is implemented by inserting a dropout layer into the all-connected and output layers, to minimize cross entropy. This study used the datasets MNIST, HCL2000, and EnglishHand as the benchmark data, analyzed the performance of the SGD optimizer under different learning parameters, and found that the proposed algorithm exhibited good recognition performance when the learning rate was set to [0.05, 0.07]. The performances of WCNN, MLP-CNN, SVM-ELM, and MCNN-DS were compared. Statistical results showed the following: (1 For the benchmark MNIST, the MCNN-DS exhibited a high recognition rate of 99.97%, and the time-cost of the proposed algorithm was merely 21.95% of MLP-CNN, and 10.02% of SVM-ELM; (2 Compared with SVM-ELM, the average improvement in the recognition rate of MCNN-DS was 2.35% for the benchmark HCL2000, and the time-cost of MCNN-DS was only 15.41%; (3 For the EnglishHand test set, the lowest recognition rate of the algorithm was 84.93%, the highest recognition rate was 95.29%, and the average recognition rate was 89.77%.

Network reconfiguration and neuronal plasticity in rhythm-generating networks.

Science.gov (United States)

Koch, Henner; Garcia, Alfredo J; Ramirez, Jan-Marino

2011-12-01

Neuronal networks are highly plastic and reconfigure in a state-dependent manner. The plasticity at the network level emerges through multiple intrinsic and synaptic membrane properties that imbue neurons and their interactions with numerous nonlinear properties. These properties are continuously regulated by neuromodulators and homeostatic mechanisms that are critical to maintain not only network stability and also adapt networks in a short- and long-term manner to changes in behavioral, developmental, metabolic, and environmental conditions. This review provides concrete examples from neuronal networks in invertebrates and vertebrates, and illustrates that the concepts and rules that govern neuronal networks and behaviors are universal.

Adaptive Neural Network Control for Nonlinear Hydraulic Servo-System with Time-Varying State Constraints

Directory of Open Access Journals (Sweden)

Shu-Min Lu

2017-01-01

Full Text Available An adaptive neural network control problem is addressed for a class of nonlinear hydraulic servo-systems with time-varying state constraints. In view of the low precision problem of the traditional hydraulic servo-system which is caused by the tracking errors surpassing appropriate bound, the previous works have shown that the constraint for the system is a good way to solve the low precision problem. Meanwhile, compared with constant constraints, the time-varying state constraints are more general in the actual systems. Therefore, when the states of the system are forced to obey bounded time-varying constraint conditions, the high precision tracking performance of the system can be easily realized. In order to achieve this goal, the time-varying barrier Lyapunov function (TVBLF is used to prevent the states from violating time-varying constraints. By the backstepping design, the adaptive controller will be obtained. A radial basis function neural network (RBFNN is used to estimate the uncertainties. Based on analyzing the stability of the hydraulic servo-system, we show that the error signals are bounded in the compacts sets; the time-varying state constrains are never violated and all singles of the hydraulic servo-system are bounded. The simulation and experimental results show that the tracking accuracy of system is improved and the controller has fast tracking ability and strong robustness.

Cooperative learning neural network output feedback control of uncertain nonlinear multi-agent systems under directed topologies

Science.gov (United States)

Wang, W.; Wang, D.; Peng, Z. H.

2017-09-01

Without assuming that the communication topologies among the neural network (NN) weights are to be undirected and the states of each agent are measurable, the cooperative learning NN output feedback control is addressed for uncertain nonlinear multi-agent systems with identical structures in strict-feedback form. By establishing directed communication topologies among NN weights to share their learned knowledge, NNs with cooperative learning laws are employed to identify the uncertainties. By designing NN-based κ-filter observers to estimate the unmeasurable states, a new cooperative learning output feedback control scheme is proposed to guarantee that the system outputs can track nonidentical reference signals with bounded tracking errors. A simulation example is given to demonstrate the effectiveness of the theoretical results.

Mobile robot nonlinear feedback control based on Elman neural network observer

Directory of Open Access Journals (Sweden)

Khaled Al-Mutib

2015-12-01

Full Text Available This article presents a new approach to control a wheeled mobile robot without velocity measurement. The controller developed is based on kinematic model as well as dynamics model to take into account parameters of dynamics. These parameters related to dynamic equations are identified using a proposed methodology. Input–output feedback linearization is considered with a slight modification in the mathematical expressions to implement the dynamic controller and analyze the nonlinear internal behavior. The developed controllers require sensors to obtain the states needed for the closed-loop system. However, some states may not be available due to the absence of the sensors because of the cost, the weight limitation, reliability, induction of errors, failure, and so on. Particularly, for the velocity measurements, the required accuracy may not be achieved in practical applications due to the existence of significant errors induced by stochastic or cyclical noise. In this article, Elman neural network is proposed to work as an observer to estimate the velocity needed to complete the full state required for the closed-loop control and account for all the disturbances and model parameter uncertainties. Different simulations are carried out to demonstrate the feasibility of the approach in tracking different reference trajectories in comparison with other paradigms.

Harmonics: Generation and Suppression in AC System Networks ...

African Journals Online (AJOL)

However, reactive power flow in electrical networks has adverse effects depending on their magnitude and the nature of the supply network. How these harmonics are generated by nonlinear loads and the means by which they can be kept low are the focus of this paper. Keywords: non-linear loads, harmonics, reactive ...

Non-linear multivariable predictive control of an alcoholic fermentation process using functional link networks

Directory of Open Access Journals (Sweden)

Luiz Augusto da Cruz Meleiro

2005-06-01

Full Text Available In this work a MIMO non-linear predictive controller was developed for an extractive alcoholic fermentation process. The internal model of the controller was represented by two MISO Functional Link Networks (FLNs, identified using simulated data generated from a deterministic mathematical model whose kinetic parameters were determined experimentally. The FLN structure presents as advantages fast training and guaranteed convergence, since the estimation of the weights is a linear optimization problem. Besides, the elimination of non-significant weights generates parsimonious models, which allows for fast execution in an MPC-based algorithm. The proposed algorithm showed good potential in identification and control of non-linear processes.Neste trabalho um controlador preditivo não linear multivariável foi desenvolvido para um processo de fermentação alcoólica extrativa. O modelo interno do controlador foi representado por duas redes do tipo Functional Link (FLN, identificadas usando dados de simulação gerados a partir de um modelo validado experimentalmente. A estrutura FLN apresenta como vantagem o treinamento rápido e convergência garantida, já que a estimação dos seus pesos é um problema de otimização linear. Além disso, a eliminação de pesos não significativos gera modelos parsimoniosos, o que permite a rápida execução em algoritmos de controle preditivo baseado em modelo. Os resultados mostram que o algoritmo proposto tem grande potencial para identificação e controle de processos não lineares.

Deep architecture neural network-based real-time image processing for image-guided radiotherapy.

Science.gov (United States)

Mori, Shinichiro

2017-08-01

To develop real-time image processing for image-guided radiotherapy, we evaluated several neural network models for use with different imaging modalities, including X-ray fluoroscopic image denoising. Setup images of prostate cancer patients were acquired with two oblique X-ray fluoroscopic units. Two types of residual network were designed: a convolutional autoencoder (rCAE) and a convolutional neural network (rCNN). We changed the convolutional kernel size and number of convolutional layers for both networks, and the number of pooling and upsampling layers for rCAE. The ground-truth image was applied to the contrast-limited adaptive histogram equalization (CLAHE) method of image processing. Network models were trained to keep the quality of the output image close to that of the ground-truth image from the input image without image processing. For image denoising evaluation, noisy input images were used for the training. More than 6 convolutional layers with convolutional kernels >5×5 improved image quality. However, this did not allow real-time imaging. After applying a pair of pooling and upsampling layers to both networks, rCAEs with >3 convolutions each and rCNNs with >12 convolutions with a pair of pooling and upsampling layers achieved real-time processing at 30 frames per second (fps) with acceptable image quality. Use of our suggested network achieved real-time image processing for contrast enhancement and image denoising by the use of a conventional modern personal computer. Copyright © 2017 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.

Multi-Level and Multi-Scale Feature Aggregation Using Pretrained Convolutional Neural Networks for Music Auto-Tagging

Science.gov (United States)

Lee, Jongpil; Nam, Juhan

2017-08-01

Music auto-tagging is often handled in a similar manner to image classification by regarding the 2D audio spectrogram as image data. However, music auto-tagging is distinguished from image classification in that the tags are highly diverse and have different levels of abstractions. Considering this issue, we propose a convolutional neural networks (CNN)-based architecture that embraces multi-level and multi-scaled features. The architecture is trained in three steps. First, we conduct supervised feature learning to capture local audio features using a set of CNNs with different input sizes. Second, we extract audio features from each layer of the pre-trained convolutional networks separately and aggregate them altogether given a long audio clip. Finally, we put them into fully-connected networks and make final predictions of the tags. Our experiments show that using the combination of multi-level and multi-scale features is highly effective in music auto-tagging and the proposed method outperforms previous state-of-the-arts on the MagnaTagATune dataset and the Million Song Dataset. We further show that the proposed architecture is useful in transfer learning.

Local stability and Hopf bifurcation in small-world delayed networks

International Nuclear Information System (INIS)

Li Chunguang; Chen Guanrong

2004-01-01

The notion of small-world networks, recently introduced by Watts and Strogatz, has attracted increasing interest in studying the interesting properties of complex networks. Notice that, a signal or influence travelling on a small-world network often is associated with time-delay features, which are very common in biological and physical networks. Also, the interactions within nodes in a small-world network are often nonlinear. In this paper, we consider a small-world networks model with nonlinear interactions and time delays, which was recently considered by Yang. By choosing the nonlinear interaction strength as a bifurcation parameter, we prove that Hopf bifurcation occurs. We determine the stability of the bifurcating periodic solutions and the direction of the Hopf bifurcation by applying the normal form theory and the center manifold theorem. Finally, we show a numerical example to verify the theoretical analysis

Local stability and Hopf bifurcation in small-world delayed networks

Energy Technology Data Exchange (ETDEWEB)

Li Chunguang E-mail: cgli@uestc.edu.cn; Chen Guanrong E-mail: gchen@ee.cityu.edu.hk

2004-04-01

The notion of small-world networks, recently introduced by Watts and Strogatz, has attracted increasing interest in studying the interesting properties of complex networks. Notice that, a signal or influence travelling on a small-world network often is associated with time-delay features, which are very common in biological and physical networks. Also, the interactions within nodes in a small-world network are often nonlinear. In this paper, we consider a small-world networks model with nonlinear interactions and time delays, which was recently considered by Yang. By choosing the nonlinear interaction strength as a bifurcation parameter, we prove that Hopf bifurcation occurs. We determine the stability of the bifurcating periodic solutions and the direction of the Hopf bifurcation by applying the normal form theory and the center manifold theorem. Finally, we show a numerical example to verify the theoretical analysis.

Modeling, Optimization & Control of Hydraulic Networks

DEFF Research Database (Denmark)

Tahavori, Maryamsadat

2014-01-01

. The nonlinear network model is derived based on the circuit theory. A suitable projection is used to reduce the state vector and to express the model in standard state-space form. Then, the controllability of nonlinear nonaffine hydraulic networks is studied. The Lie algebra-based controllability matrix is used......Water supply systems consist of a number of pumping stations, which deliver water to the customers via pipeline networks and elevated reservoirs. A huge amount of drinking water is lost before it reaches to end-users due to the leakage in pipe networks. A cost effective solution to reduce leakage...... in water network is pressure management. By reducing the pressure in the water network, the leakage can be reduced significantly. Also it reduces the amount of energy consumption in water networks. The primary purpose of this work is to develop control algorithms for pressure control in water supply...

A Nonlinear Programming and Artificial Neural Network Approach for Optimizing the Performance of a Job Dispatching Rule in a Wafer Fabrication Factory

Directory of Open Access Journals (Sweden)

Toly Chen

2012-01-01

Full Text Available A nonlinear programming and artificial neural network approach is presented in this study to optimize the performance of a job dispatching rule in a wafer fabrication factory. The proposed methodology fuses two existing rules and constructs a nonlinear programming model to choose the best values of parameters in the two rules by dynamically maximizing the standard deviation of the slack, which has been shown to benefit scheduling performance by several studies. In addition, a more effective approach is also applied to estimate the remaining cycle time of a job, which is empirically shown to be conducive to the scheduling performance. The efficacy of the proposed methodology was validated with a simulated case; evidence was found to support its effectiveness. We also suggested several directions in which it can be exploited in the future.

Learning scale-variant and scale-invariant features for deep image classification

NARCIS (Netherlands)

van Noord, Nanne; Postma, Eric

Convolutional Neural Networks (CNNs) require large image corpora to be trained on classification tasks. The variation in image resolutions, sizes of objects and patterns depicted, and image scales, hampers CNN training and performance, because the task-relevant information varies over spatial

All-optical and digital non-linear compensation algorithms in flex-coherent grouped and un-grouped contiguous spectrum based networks

DEFF Research Database (Denmark)

Asif, Rameez

2016-01-01

We have evaluated that in-line non-linear compensation schemes decrease the complexity of digital backward propagation and enhance the transmission performance of 40/112/224 Gbit/s mixed line rate network. Multiple bit rates, i.e. 40/112/224 Gbit/s and modulation formats (i.e. DP-QPSK and DP-16QAM......) are transmitted over 1280 km of Large $$\\hbox {A}_{eff}$$ A e f f Pure-Silica core fiber. Both grouped and un-grouped spectral allocation schemes are investigated. Optical add-drop multiplexers are used to drop the required wavelength for signal processing in the transmission link. Moreover, hybrid mid...

Nonlinear Maps and their Applications 2011 International Workshop

CERN Document Server

Fournier-Prunaret, Daniele; Ueta, Tetsushi; Nishio, Yoshifumi

2014-01-01

In the field of Dynamical Systems, nonlinear iterative processes play an important role. Nonlinear mappings can be found as immediate models for many systems from different scientific areas, such as engineering, economics, biology, or can also be obtained via numerical methods permitting to solve non-linear differential equations. In both cases, the understanding of specific dynamical behaviors and phenomena is of the greatest interest for scientists. This volume contains papers that were presented at the International Workshop on Nonlinear Maps and their Applications (NOMA 2011) held in Évora, Portugal, on September 15-16, 2011. This kind of collaborative effort is of paramount importance in promoting communication among the various groups that work in dynamical systems and networks in their research theoretical studies as well as for applications. This volume is suitable for graduate students as well as researchers in the field.

Convolutional neural networks with balanced batches for facial expressions recognition

Science.gov (United States)

Battini Sönmez, Elena; Cangelosi, Angelo

2017-03-01

This paper considers the issue of fully automatic emotion classification on 2D faces. In spite of the great effort done in recent years, traditional machine learning approaches based on hand-crafted feature extraction followed by the classification stage failed to develop a real-time automatic facial expression recognition system. The proposed architecture uses Convolutional Neural Networks (CNN), which are built as a collection of interconnected processing elements to simulate the brain of human beings. The basic idea of CNNs is to learn a hierarchical representation of the input data, which results in a better classification performance. In this work we present a block-based CNN algorithm, which uses noise, as data augmentation technique, and builds batches with a balanced number of samples per class. The proposed architecture is a very simple yet powerful CNN, which can yield state-of-the-art accuracy on the very competitive benchmark algorithm of the Extended Cohn Kanade database.

Classification and Segmentation of Satellite Orthoimagery Using Convolutional Neural Networks

Directory of Open Access Journals (Sweden)

Martin Längkvist

2016-04-01

Full Text Available The availability of high-resolution remote sensing (HRRS data has opened up the possibility for new interesting applications, such as per-pixel classification of individual objects in greater detail. This paper shows how a convolutional neural network (CNN can be applied to multispectral orthoimagery and a digital surface model (DSM of a small city for a full, fast and accurate per-pixel classification. The predicted low-level pixel classes are then used to improve the high-level segmentation. Various design choices of the CNN architecture are evaluated and analyzed. The investigated land area is fully manually labeled into five categories (vegetation, ground, roads, buildings and water, and the classification accuracy is compared to other per-pixel classification works on other land areas that have a similar choice of categories. The results of the full classification and segmentation on selected segments of the map show that CNNs are a viable tool for solving both the segmentation and object recognition task for remote sensing data.

Neural Networks in Control Applications

DEFF Research Database (Denmark)

Sørensen, O.

are examined. The models are separated into three groups representing input/output descriptions as well as state space descriptions: - Models, where all in- and outputs are measurable (static networks). - Models, where some inputs are non-measurable (recurrent networks). - Models, where some in- and some...... outputs are non-measurable (recurrent networks with incomplete state information). The three groups are ordered in increasing complexity, and for each group it is shown how to solve the problems concerning training and application of the specific model type. Of particular interest are the model types...... Kalmann filter) representing state space description. The potentials of neural networks for control of non-linear processes are also examined, focusing on three different groups of control concepts, all considered as generalizations of known linear control concepts to handle also non-linear processes...

Nonlinear PCA: characterizing interactions between modes of brain activity.

OpenAIRE

Friston, K; Phillips, J; Chawla, D; Büchel, C

2000-01-01

This paper presents a nonlinear principal component analysis (PCA) that identifies underlying sources causing the expression of spatial modes or patterns of activity in neuroimaging time-series. The critical aspect of this technique is that, in relation to conventional PCA, the sources can interact to produce (second-order) spatial modes that represent the modulation of one (first-order) spatial mode by another. This nonlinear PCA uses a simple neural network architecture that embodies a spec...

Identifying time-delayed gene regulatory networks via an evolvable hierarchical recurrent neural network.

Science.gov (United States)

Kordmahalleh, Mina Moradi; Sefidmazgi, Mohammad Gorji; Harrison, Scott H; Homaifar, Abdollah

2017-01-01

The modeling of genetic interactions within a cell is crucial for a basic understanding of physiology and for applied areas such as drug design. Interactions in gene regulatory networks (GRNs) include effects of transcription factors, repressors, small metabolites, and microRNA species. In addition, the effects of regulatory interactions are not always simultaneous, but can occur after a finite time delay, or as a combined outcome of simultaneous and time delayed interactions. Powerful biotechnologies have been rapidly and successfully measuring levels of genetic expression to illuminate different states of biological systems. This has led to an ensuing challenge to improve the identification of specific regulatory mechanisms through regulatory network reconstructions. Solutions to this challenge will ultimately help to spur forward efforts based on the usage of regulatory network reconstructions in systems biology applications. We have developed a hierarchical recurrent neural network (HRNN) that identifies time-delayed gene interactions using time-course data. A customized genetic algorithm (GA) was used to optimize hierarchical connectivity of regulatory genes and a target gene. The proposed design provides a non-fully connected network with the flexibility of using recurrent connections inside the network. These features and the non-linearity of the HRNN facilitate the process of identifying temporal patterns of a GRN. Our HRNN method was implemented with the Python language. It was first evaluated on simulated data representing linear and nonlinear time-delayed gene-gene interaction models across a range of network sizes and variances of noise. We then further demonstrated the capability of our method in reconstructing GRNs of the Saccharomyces cerevisiae synthetic network for in vivo benchmarking of reverse-engineering and modeling approaches (IRMA). We compared the performance of our method to TD-ARACNE, HCC-CLINDE, TSNI and ebdbNet across different network

Using function approximation to determine neural network accuracy

International Nuclear Information System (INIS)