Showing papers in "Neurocomputing in 2010"

TL;DR: Under the ELM learning framework, SVM's maximal margin property and the minimal norm of weights theory of feedforward neural networks are actually consistent and ELM for classification tends to achieve better generalization performance than traditional SVM.

TL;DR: This article presents a comprehensive overview of the hardware realizations of artificial neural network models, known as hardware neural networks (HNN), appearing in academic studies as prototypes as well as in commercial use.

TL;DR: Numerical experimental results confirm that the proposed method can predict the chaotic time series more effectively and accurately when compared with the existing prediction methods.

TL;DR: This work identifies an extensive feature set describing both the time series and the pool of individual forecasting methods and investigates the applicability of different meta-learning approaches, showing the superiority of a ranking-based combination of methods over simple model selection approaches.

TL;DR: In this paper, a review of rule extraction from SVM classifiers is presented, and a comparison of the algorithms' salient features and relative performance as measured by a number of metrics is made.

TL;DR: The aim of using correlation information in CAFS is to encourage the search strategy for selecting less correlated features if they enhance accuracy of NNs and reduce redundancy of information resulting in compact NN architectures.

TL;DR: The experimental section shows that multiple- Output approaches represent a competitive choice for tackling long-term forecasting tasks and goes a step forward with respect to the previous authors contributions by extending the multiple-output approach with a query-based criterion.

TL;DR: This work proposes a fully data driven forecasting methodology that combines filter and wrapper approaches for feature selection, including automatic feature evaluation, construction and transformation for heterogeneous sets of time series without expert intervention.

TL;DR: Based on a large number of experiments, the NGHS has demonstrated stronger convergence and stability than original harmony search (HS) algorithm and its two improved algorithms (IHS and SGHS).

TL;DR: This article constitutes the first half of a two-part world survey of artificial brain projects: this part dealing with large-scale brain simulations, and the second part with biologically inspired cognitive architectures (BICAs).

TL;DR: A novel recognition framework for human actions using hybrid features extracted using motion-selectivity attribute of 3D dual-tree complex wavelet transform and affine SIFT local image detector which offers enhanced capabilities to preserve structure and correlation amongst neighborhood pixels of a video frame.

TL;DR: A simple and systematic approach is developed for modeling and neural adaptive backstepping control of an uncertain chaotic system, using only input-output data obtained from the underlying dynamical systems.

TL;DR: The proposed constructive hidden nodes selection for ELM (referred to as CS-ELM) selects the optimal number of hidden nodes when the unbiased risk estimation based criterion C"P reaches the minimum value.

TL;DR: A number of leading cognitive architectures that are inspired by the human brain, at various levels of granularity, are reviewed and compared, with special attention paid to the way their internal structures and dynamics map onto neural processes.

TL;DR: It is shown that a neural network based model using a set of variables selected with a criterion that it is adapted to the network leads to better results than a set chosen with criteria used in the financial literature.

TL;DR: The proposed model of neutral type is quite general since many factors such as noise perturbations, Markovian jump parameters and mixed time delays are considered in this paper and they are more general than those usual Lipschitz conditions.

TL;DR: The approach makes use of self-adaptive error based control parameters to alter the training data sequence, evolve the network architecture, and learn the network parameters and avoids the over-training problem and reduces the training time significantly.

TL;DR: Lyapunov-Krasovskii functional combining with the input delay approach as well as the improved free-weighting matrix approach are employed to derive several sufficient criteria ensuring the delayed neural networks to be exponentially synchronous.

TL;DR: The proposed LPDP combining manifold criterion and Fisher criterion has more discriminanting power, and is more suitable for recognition tasks than LPP, which considers only the local information for classification tasks.

TL;DR: A sufficient condition is derived to ensure the global synchronization of switched linearly coupled complex neural networks, which are controlled by some designed controllers, and a globally convergent algorithm involving convex optimization is presented to construct such controllers effectively.

TL;DR: By constructing a novel Lyapunov functional and introducing some appropriate free-weighting matrices, delay-dependent passivity analysis criteria are derived and can be developed in the frame of convex optimization problems and then solved via standard numerical software.

TL;DR: A tighter upper bound of the differential of Lyapunov-Krasovskii functional is obtained by an improved approximation method and some delay-dependent passivity criteria are obtained in the linear matrix inequality (LMI) format.

TL;DR: This paper gives a rigorous justification to reveal that under some mild conditions, the kernelization under this framework is equivalent to traditional kernel method, and shows that these mild conditions are usually satisfied in most of learning algorithms.

TL;DR: A systematic two-stage algorithm (named TS-ELM) is introduced to handle the problem and drastically reduces the network complexity of extreme learning machine (ELM).

TL;DR: The state estimation problem for discrete neural networks with Markovian jumping parameters and time-varying delays is investigated and the delay-dependent sufficient conditions for the existence of desired state estimator are derived.

TL;DR: The robust H"~ state estimation problem is investigated for a general class of uncertain discrete-time stochastic neural networks with probabilistic measurement delays and the explicit expression of the desired estimator gains is described in terms of the solution to a linear matrix inequality (LMI).

TL;DR: In the proposed wavelet neural networks, composite functions are applied at the hidden nodes and the learning is done using ELM, which can achieve better performances in most cases than some relevant neural networks and learn much faster than neural networks training with the traditional back-propagation (BP) algorithm.

TL;DR: The synchronization problem of continuous/discrete general complex dynamical networks with time-varying delays with some new delay-dependent synchronization stability criteria derived in the form of linear matrix inequalities are investigated.

TL;DR: A novel Lyapunov functional dependence on auxiliary delay parameters is exploited, which renders the results to be potentially less conservative and allows the time-varying delays to be not differentiable.