Showing papers in "Neurocomputing in 2012"

TL;DR: This work employs two problem decomposition methods for training Elman recurrent neural networks on chaotic time series problems and shows improvement in performance in terms of accuracy when compared to some of the methods from literature.

TL;DR: This paper presents the detailed design architecture and its associated learning algorithm to explain how effective learning and optimization can be achieved in this new ADP architecture and test the performance both on the cart-pole balancing task and the triple-link inverted pendulum balancing task.

TL;DR: A finite-horizon neuro-optimal tracking control strategy for a class of discrete-time nonlinear systems and three neural networks are used as parametric structures to implement the algorithm, which aims at approximating the cost function, the control law, and the error dynamics.

TL;DR: It is demonstrated that the aa model is relevant for feature extraction and dimensionality reduction for a large variety of machine learning problems taken from computer vision, neuroimaging, chemistry, text mining and collaborative filtering leading to highly interpretable representations of the dynamics in the data.

TL;DR: It can be found that the fractional-order four-cell cellular neural network proposed and investigated by means of numerical simulations does exhibit hyperchaotic phenomena over a wide range of values of some specified parameter.

TL;DR: The experimental results show that FOS-ELM has higher accuracy with fewer training time, better stability and short-term predictability than EOS- ELM.

TL;DR: These results ensure global exponential stability of memristor-based neural networks in the sense of Filippov solutions, and it is convenient to estimate the exponential convergence rates of this neural network by using the results.

TL;DR: This paper proposes a framework for approximate NMF which constrains the ℓ0-norm of the basis matrix, or the coefficient matrix, respectively, and demonstrates the benefits of these methods, which compare to, or outperform existing approaches.

TL;DR: This paper develops an algorithm capable of determining the step-changes in signals that occur whenever a device is turned on or off, and which allows for the definition of a unique signature (ID) for each device.

TL;DR: The proposed Hybrid Artificial Bee Colony (HABC) algorithm is proved to have significant improvement over canonical ABC and several other comparison algorithms and is a competitive approach for data clustering.

TL;DR: A kernel sparse representation based classification (KSRC) algorithm is proposed, which has more powerful classification ability than SRC and is demonstrated to be effective in face recognition, palmprint recognition and finger-knuckle-print recognition.

TL;DR: The proposed algorithm makes full use of the affine invariant advantage of ASIFT and the efficient merit of SURF while avoids their drawbacks and demonstrates the robustness and efficiency of the proposed algorithm.

TL;DR: An mth order nonlinear model to describe the relationship between the surface electromyography (sEMG) signals and the joint angles of human legs is proposed, in which a simple BP neural network is built for the model estimation.

TL;DR: The global stability of the proposed neural network and the optimality of the neural solution are proven in theory and application orientated simulations demonstrate the effectiveness of this proposed method.

TL;DR: This work investigates the combination of meta-learning with two search algorithms: Particle Swarm Optimization and Tabu Search and showed that, on average, the proposed hybrid methods obtained lower error rates when compared to their components applied in isolation.

TL;DR: The learning ability of neural networks is used to design a robust adaptive backstepping controller that does not require the knowledge of the robot dynamics and gains are tuned on-line to minimize the velocity error and improve the trajectory tracking characteristics.

TL;DR: A sequential learning algorithm for a neural network classifier based on human meta-cognitive learning principles, which indicates the superior performance of McNN over reported results in the literature.

TL;DR: This paper addresses the design of sliding mode controller (SMC) for an uncertain chaotic fractional order economic system and an adaptive SMC is designed in the case that the upper bound of the uncertainties is unknown.

TL;DR: The dynamic analysis in the paper employs results from the theory of differential equations with discontinuous right-hand side as introduced by Filippov, and some new conditions concerning global exponential stability are obtained.

TL;DR: Experimental results show that the TGCA has derived better performance on the search of the cluster numbers and higher accuracy on clustering problems.

TL;DR: Experimental results show that the semi-supervised multi-dimensional approach outperforms the most common Sentiment Analysis approaches, and it is concluded that this approach is beneficial to improve the recognition rates for this problem, and in extension, could be considered to solve future Sentiment analysis problems.

TL;DR: The divergence which measures the difference between probability distributions in the original and the embedding space can be treated independently from other components like, e.g. the similarity of data points or the data distribution.

TL;DR: Experimental results show that, compared with NCA, FNCA not only significantly increases the training speed but also obtains higher classification accuracy, and comparative studies with the state-of-the-art approaches on various real-world datasets also verify the effectiveness of the proposed linear and nonlinear F NCA methods.

TL;DR: A subject transfer framework for EEG classification that can achieve positive knowledge transfer for improving the performance of EEG classification when the training set of the target subject is small owing to the need to reduce the calibration session is proposed.

TL;DR: This paper uses textual information to aid the financial time series forecasting and presents a novel text mining approach via combining ARIMA and SVR (Support Vector Regression) to forecasting.

TL;DR: Experimental results show that the WNN algorithm can remove EEG artifacts effectively without diminishing useful EEG information even for very noisy datasets.

TL;DR: A novel gender classification framework, which utilizes not only facial features, but also external information, i.e. hair and clothing, which improves classification accuracy, even when images contain occlusions, noise, and illumination changes.

TL;DR: The proposed supervised sparse representation method for face recognition can achieve promising classification accuracy and is exploited as a heuristic strategy to achieve this goal.

TL;DR: The CPSFC algorithm utilizes CPSO to search the fuzzy clustering model, exploiting the searching capability of fuzzy c-means (FCM) and avoiding its major limitation of getting stuck at locally optimal values.

TL;DR: In this paper, a novel fuzzy adaptive controller is investigated for a class of multi-input multi-output (MIMO) nonaffine systems with unknown control direction by incorporating in the control law a Nussbaum-type function.