<![CDATA[ IEEE Transactions on Neural Networks and Learning Systems - new TOC ]]>
http://ieeexplore.ieee.org
TOC Alert for Publication# 5962385 2018April 23<![CDATA[Table of contents]]>295C11395125<![CDATA[IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS publication information]]>295C2C267<![CDATA[Logistic Localized Modeling of the Sample Space for Feature Selection and Classification]]>localized feature selection for which each region of the sample space is characterized by its individual distinct feature subset that may vary in size and membership. This approach can therefore select an optimal feature subset that adapts to local variations of the sample space, and hence offer the potential for improved performance. Feature subsets are computed by choosing an optimal coordinate space so that, within a localized region, within-class distances and between-class distances are, respectively, minimized and maximized. Distances are measured using a logistic function metric within the corresponding region. This enables the optimization process to focus on a localized region within the sample space. A local classification approach is utilized for measuring the similarity of a new input data point to each class. The proposed logistic localized feature selection (lLFS) algorithm is invariant to the underlying probability distribution of the data; hence, it is appropriate when the data are distributed on a nonlinear or disjoint manifold. lLFS is efficiently formulated as a joint convex/increasing quasi-convex optimization problem with a unique global optimum point. The method is most applicable when the number of available training samples is small. The performance of the proposed localized method is successfully demonstrated on a large variety of data sets. We demonstrate that the number of features selected by the lLFS method saturates at the number of available discriminative features. In addition, we have shown that the Vapnik-Chervonenkis dimension of the localized classifier is finite. Both these factors suggest that the lLFS method is insensitive to the overfitting issue, relative to other methods.]]>295139614132931<![CDATA[Manifold Warp Segmentation of Human Action]]>295141414263614<![CDATA[Computational Model Based on Neural Network of Visual Cortex for Human Action Recognition]]>295142714404258<![CDATA[DeepX: Deep Learning Accelerator for Restricted Boltzmann Machine Artificial Neural Networks]]>$1024times 1024$ network size, using 128 input cases per batch, and running at a 303-MHz clock frequency) integrated in a state-of-the art field-programmable gate array (FPGA) (Xilinx Virtex 7 XC7V-2000T) provides a computational performance of 301-billion connection-updates-per-second and about 193 times higher performance than a software solution running on general purpose processors. Most importantly, the architecture enables over 4 times (12 times in batch learning) higher performance compared with a previous work when both are implemented in an FPGA device (XC2VP70).]]>295144114531995<![CDATA[Preconditioned Stochastic Gradient Descent]]>295145414661955<![CDATA[Global <inline-formula> <tex-math notation="LaTeX">$H_infty $ </tex-math></inline-formula> Pinning Synchronization of Complex Networks With Sampled-Data Communications]]>$H_infty $ pinning synchronization problem for a class of complex networks with aperiodic samplings. Combined with the Writinger-based integral inequality, a new less conservative criterion is presented to guarantee the global pinning synchronization of the complex network. Furthermore, a novel condition is proposed under which the complex network is globally pinning synchronized with a given $H_infty $ performance index. It is shown that the $H_infty $ performance index $gamma $ has a positive correlation with the upper bound of the sampling intervals. Finally, the validity and the advantage of the theoretic results obtained are verified by means of the applications in Chua’s circuit and pendulum.]]>295146714764168<![CDATA[Robust Finite-Time Stabilization of Fractional-Order Neural Networks With Discontinuous and Continuous Activation Functions Under Uncertainty]]>295147714901346<![CDATA[Stability Analysis and Application for Delayed Neural Networks Driven by Fractional Brownian Noise]]>295149115021349<![CDATA[Discriminative Sparse Neighbor Approximation for Imbalanced Learning]]>295150315132723<![CDATA[Data-Driven Multiagent Systems Consensus Tracking Using Model Free Adaptive Control]]>295151415241797<![CDATA[Reinforced Robust Principal Component Pursuit]]>295152515382330<![CDATA[Adaptive Boundary Iterative Learning Control for an Euler–Bernoulli Beam System With Input Constraint]]>295153915491557<![CDATA[Synchronization of Coupled Reaction–Diffusion Neural Networks With Directed Topology via an Adaptive Approach]]>295155015611051<![CDATA[Solving Multiextremal Problems by Using Recurrent Neural Networks]]>295156215742520<![CDATA[Multitarget Sparse Latent Regression]]>$ell _{2,1}$ -norm-based sparse learning; the MSLR naturally admits a representer theorem for kernel extension, which enables it to flexibly handle highly complex nonlinear input-output relationships; the MSLR can be solved efficiently by an alternating optimization algorithm with guaranteed convergence, which ensures efficient multitarget regression. Extensive experimental evaluation on both synthetic data and six greatly diverse real-world data sets shows that the proposed MSLR consistently outperforms the state-of-the-art algorithms, which demonstrates its great effectiveness for multivariate prediction.]]>295157515861500<![CDATA[Convolution in Convolution for Network in Network]]>$ 1times 1 $ convolutions in spatial domain, NiN has stronger ability of feature representation and hence results in better recognition performance. However, MLP itself consists of fully connected layers that give rise to a large number of parameters. In this paper, we propose to replace dense shallow MLP with sparse shallow MLP. One or more layers of the sparse shallow MLP are sparely connected in the channel dimension or channel–spatial domain. The proposed method is implemented by applying unshared convolution across the channel dimension and applying shared convolution across the spatial dimension in some computational layers. The proposed method is called convolution in convolution (CiC). The experimental results on the CIFAR10 data set, augmented CIFAR10 data set, and CIFAR100 data set demonstrate the effectiveness of the proposed CiC method.]]>295158715972493<![CDATA[Application of LMS-Based NN Structure for Power Quality Enhancement in a Distribution Network Under Abnormal Conditions]]>295159816072642<![CDATA[A Confident Information First Principle for Parameter Reduction and Model Selection of Boltzmann Machines]]>295160816211480<![CDATA[AnRAD: A Neuromorphic Anomaly Detection Framework for Massive Concurrent Data Streams]]>295162216362908<![CDATA[Dynamic Uncertain Causality Graph for Knowledge Representation and Reasoning: Utilization of Statistical Data and Domain Knowledge in Complex Cases]]>$i$ -mode, $e$ -mode, and $h$ -mode of the DUCG to model such complex cases and then transform them into either the standard $i$ -mode or the standard $e$ -mode. In the former situation, if no directed cyclic graph is involved, the transformed result is simply a Bayesian network (BN), and existing inference methods for BNs can be applied. In the latter situation, an inference method based on the DUCG is proposed. Examples are provided to illustrate the methodology.]]>295163716513285<![CDATA[Recurrent Neural Networks With Auxiliary Memory Units]]>295165216611909<![CDATA[Random Forest Classifier for Zero-Shot Learning Based on Relative Attribute]]>295166216744043<![CDATA[Improving Crowdsourced Label Quality Using Noise Correction]]>295167516884266<![CDATA[Hyperbolic Gradient Operator and Hyperbolic Back-Propagation Learning Algorithms]]>295168917021533<![CDATA[Autonomous Data Collection Using a Self-Organizing Map]]>295170317152562<![CDATA[A Preference-Based Multiobjective Evolutionary Approach for Sparse Optimization]]>$k$ -sparse solution. However, the setting of regularization parameters or the estimation of the true sparsity are nontrivial in iterative thresholding methods. To overcome this shortcoming, we propose a preference-based multiobjective evolutionary approach to solve sparse optimization problems in compressive sensing. Our basic strategy is to search the knee part of weakly Pareto front with preference on the true $k$ -sparse solution. In the noiseless case, it is easy to locate the exact position of the $k$ -sparse solution from the distribution of the solutions found by our proposed method. Therefore, our method has the ability to detect the true sparsity. Moreover, any iterative thresholding methods can be used as a local optimizer in our proposed method, and no prior estimation of sparsity is required. The proposed method can also be extended to solve sparse optimization problems with noise. Extensive experiments have been conducted to study its performance on artificial signals and magnetic resonance imaging signals. Our experimental results have shown that our proposed method is very effective for detecting sparsity and can improve the reconstruction ability of existing iterative thresholding methods.]]>295171617312772<![CDATA[Asynchronous State Estimation for Discrete-Time Switched Complex Networks With Communication Constraints]]>$H_{infty }$ performance level is also ensured. The characterization of the desired estimator gains is derived in terms of the solution to a convex optimization problem. Finally, the effectiveness of the proposed design approach is demonstrated by a simulation example.]]>295173217461539<![CDATA[Cluster Synchronization for Interacting Clusters of Nonidentical Nodes via Intermittent Pinning Control]]>295174717592603<![CDATA[SCE: A Manifold Regularized Set-Covering Method for Data Partitioning]]>295176017733328<![CDATA[Efficient kNN Classification With Different Numbers of Nearest Neighbors]]>${k}$ nearest neighbor (kNN) method is a popular classification method in data mining and statistics because of its simple implementation and significant classification performance. However, it is impractical for traditional kNN methods to assign a fixed ${k}$ value (even though set by experts) to all test samples. Previous solutions assign different $k$ values to different test samples by the cross validation method but are usually time-consuming. This paper proposes a kTree method to learn different optimal $k$ values for different test/new samples, by involving a training stage in the kNN classification. Specifically, in the training stage, kTree method first learns optimal $k$ values for all training samples by a new sparse reconstruction model, and then constructs a decision tree (namely, kTree) using training samples and the learned optimal $k$ values. In the test stage, the kTree fast outputs the optimal $k$ value for each test sample, and then, the kNN classification can be conducted using the learned optimal $k$ value and all training samples. As a result, the proposed kTree method has a similar running cost but higher classification accuracy, compared with traditional kNN methods, which assign a fixed ${k}$ value to all test samples. Moreover, the proposed kTree method needs less running cost but achieves simil-
r classification accuracy, compared with the newly kNN methods, which assign different ${k}$ values to different test samples. This paper further proposes an improvement version of kTree method (namely, k*Tree method) to speed its test stage by extra storing the information of the training samples in the leaf nodes of kTree, such as the training samples located in the leaf nodes, their kNNs, and the nearest neighbor of these kNNs. We call the resulting decision tree as k*Tree, which enables to conduct kNN classification using a subset of the training samples in the leaf nodes rather than all training samples used in the newly kNN methods. This actually reduces running cost of test stage. Finally, the experimental results on 20 real data sets showed that our proposed methods (i.e., kTree and k*Tree) are much more efficient than the compared methods in terms of classification tasks.]]>295177417852775<![CDATA[Manifold Regularized Correlation Object Tracking]]>295178617951731<![CDATA[Bioinspired Approach to Modeling Retinal Ganglion Cells Using System Identification Techniques]]>295179618084438<![CDATA[Synchronization Criteria for Discontinuous Neural Networks With Mixed Delays via Functional Differential Inclusions]]>295180918211811<![CDATA[New Conditions for Global Asymptotic Stability of Memristor Neural Networks]]>charge–flux domain, instead of the typical current–voltage domain as it happens for Hopfield NNs and standard cellular NNs. One key advantage is that, when a steady state is reached, all currents, voltages, and power of a DM-NN drop off, whereas the memristors act as nonvolatile memories that store the processing result. Previous work in the literature addressed multistability of DM-NNs, i.e., convergence of solutions in the presence of multiple asymptotically stable equilibrium points (EPs). The goal of this paper is to study a basically different dynamical property of DM-NNs, namely, to thoroughly investigate the fundamental issue of global asymptotic stability (GAS) of the unique EP of a DM-NN in the general case of nonsymmetric neuron interconnections. A basic result on GAS of DM-NNs is established using Lyapunov method and the concept of Lyapunov diagonally stable matrices. On this basis, some relevant classes of nonsymmetric DM-NNs enjoying the property of GAS are highlighted.]]>29518221834998<![CDATA[Doubly Nonparametric Sparse Nonnegative Matrix Factorization Based on Dependent Indian Buffet Processes]]>295183518491804<![CDATA[Event-Sampled Direct Adaptive NN Output- and State-Feedback Control of Uncertain Strict-Feedback System]]>295185018632030<![CDATA[Modeling and Analysis of Beta Oscillations in the Basal Ganglia]]>295186418752603<![CDATA[Safe Screening Rules for Accelerating Twin Support Vector Machine Classification]]>295187618873996<![CDATA[Dissipativity-Based Resilient Filtering of Periodic Markovian Jump Neural Networks With Quantized Measurements]]>295188818991596<![CDATA[Singularities of Three-Layered Complex-Valued Neural Networks With Split Activation Function]]>295190019071264<![CDATA[A Novel Recurrent Neural Network for Manipulator Control With Improved Noise Tolerance]]>295190819182206<![CDATA[Sensitivity Analysis for Probabilistic Neural Network Structure Reduction]]>295191919321986<![CDATA[Observer-Based Robust Coordinated Control of Multiagent Systems With Input Saturation]]>295193319461835<![CDATA[Robust Structured Nonnegative Matrix Factorization for Image Representation]]>$ell _{2,p}$ -norm (especially when $0<pleq 1$ ) loss function. Specifically, the problems of noise and outliers are well addressed by the $ell _{2,p}$ -norm ($0<pleq 1$ ) loss function, while the discriminative representations of both the labeled and unlabeled data are simultaneously learned by explicitly exploring the block-diagonal structure. The proposed problem is formulated as an optimization problem with a well-defined objective function solved by the proposed iterative algorithm. The convergence of the proposed optimization algorithm is analyzed both theoretically and empirically. In addition, we also discuss the relationships between the proposed method and some previous methods. Extensive experiments on both the synthetic and real-world data sets are conducted, and the experimental results demonstrate the effectiveness of the proposed method in comparison to the state-of-the-art methods.]]>295194719602333<![CDATA[Extended Polynomial Growth Transforms for Design and Training of Generalized Support Vector Machines]]>295196119744423<![CDATA[On Better Exploring and Exploiting Task Relationships in Multitask Learning: Joint Model and Feature Learning]]>295197519851493<![CDATA[Robust Multiview Data Analysis Through Collective Low-Rank Subspace]]>295198619972022<![CDATA[Tensor-Factorized Neural Networks]]>not generalized for learning the representation from multiway observations. The classification performance using vectorized NN is constrained, because the temporal or spatial information in neighboring ways is disregarded. More parameters are required to learn the complicated data structure. This paper presents a new tensor-factorized NN (TFNN), which tightly integrates TF and NN for multiway feature extraction and classification under a unified discriminative objective. This TFNN is seen as a generalized NN, where the affine transformation in an NN is replaced by the multilinear and multiway factorization for tensor-based NN. The multiway information is preserved through layerwise factorization. Tucker decomposition and nonlinear activation are performed in each hidden layer. The tensor-factorized error backpropagation is developed to train TFNN with the limited parameter size and computation time. This TFNN can be further extended to realize the convolutional TFNN (CTFNN) by looking at small subtensors through the factorized convolution. Experiments on real-world classification tasks demonstrate that TFNN and CTFNN attain substantial improvement when compared with an NN and a convolutional NN, respectively.]]>295199820113725<![CDATA[A Novel Pruning Algorithm for Smoothing Feedforward Neural Networks Based on Group Lasso Method]]>Weight Decay, Weight Elimination, and Approximate Smoother, on both generalization and pruning efficiency. In addition, detailed simulations based on a specific data set have been performed to compare with some other common pruning strategies, which verify the advantages of the proposed algorithm. The pruning abilities of the proposed strategy have been investigated in detail for a relatively large data set, MNIST, in terms of various smoothing approximation cases.]]>295201220243900<![CDATA[Classification With Truncated <inline-formula> <tex-math notation="LaTeX">$ell _{1}$ </tex-math></inline-formula> Distance Kernel]]>$ell _{1}$ distance (TL1) kernel, which results in a classifier that is nonlinear in the global region but is linear in each subregion. With this kernel, the subregion structure can be trained using all the training data and local linear classifiers can be established simultaneously. The TL1 kernel has good adaptiveness to nonlinearity and is suitable for problems which require different nonlinearities in different areas. Though the TL1 kernel is not positive semidefinite, some classical kernel learning methods are still applicable which means that the TL1 kernel can be directly used in standard toolboxes by replacing the kernel evaluation. In numerical experiments, the TL1 kernel with a pregiven parameter achieves similar or better performance than the radial basis function kernel with the parameter tuned by cross validation, implying the TL1 kernel a promising nonlinear kernel for classification tasks.]]>295202520301089<![CDATA[Policy Iteration Algorithm for Optimal Control of Stochastic Logical Dynamical Systems]]>29520312036515<![CDATA[IEEE Computational Intelligence Society Information]]>295C3C358<![CDATA[IEEE Transactions on Neural Networks information for authors]]>295C4C4132