Neural Information Processing: 13th International Conference, ICONIP 2006, Hong Kong, China, October 3-6, 2006, Proceedings, Part I (Lecture Notes in Computer Science, 4232)

LNCS 4232 - Neural Information Processing, Part 1 A Competitive Co-evolving Support Vector Clustering Introduction Related Works Co-evolutionary Computing Support Vector Clustering A Competitive Co-evolving Support Vector Clustering Optimal Clustering Competitive Co-evolving SVC for Optimal Clustering Experimental Results Experiment 1: Accuracy Experiment 2: Training Time Conclusion References A Cooperation Online Reinforcement Learning Approach in Ant-Q Introduction Ant-TD: Ant-Q Using Temporal Difference Online Reinforcement Learning Approach in Ant-Q Evaluation Conclusion A Distributed Support Vector Machines Architecture for Chaotic Time Series Prediction Introduction Embedding Phase Space Reconstruction The Distributed Support Vector Machines Constructing the Distributed SVMs The Fuzzy Synthesis Algorithm The Proposed Algorithm for the Distributed SVMs Simulation Results Conclusions References A Divide-and-Conquer Approach to the Pairwise Opposite Class-Nearest Neighbor (POC-NN) Algorithm for Regression Problem Introduction The Methodology of POC-NN for Regression Problem Finding POC-NN Patterns Algorithm for Regression The POC-NN Algorithm for Regression Problem Approximating Algorithm by POC-NN Experimental Results Data Sets Accuracy of Approximation Evaluate the POC-NN Algorithm Conclusion A Fast Bit-Parallel Algorithm for Gapped String Kernels Introduction Gapped String Kernels Efficient Computations of Gapped Kernels Dynamic Programming Algorithm Sparse Dynamic Programming Bit-Parallel Algorithm Experiments Results on Random Strings Results on UNM Datasets Conclusions References A Heuristic Weight-Setting Algorithm for Robust Weighted Least Squares Support Vector Regression Introduction Brief Review to WLS-SVM The Proposed Weight-Setting Algorithm A Heuristic Method to Label the Outlierness of Samples The Proposed Weight-Setting Strategy Numerical Experiments Conclusions References A Hybrid Blind Signal Separation Algorithm: Particle Swarm Optimization on Feed-Forward Neural Network Introduction Blind Signal Separation Problem Description Separation Criterion PSO-Based Neural Network Blind Separation System Particle Swarm Optimization Fitness Evaluation Algorithm Pseudo Code of PSO-Based NN BSS Simulation Results and Discussion Two Sources Case Three Sources Case Discussion Conclusion A Improved SVM and Its Using in Electric Power System Load Forecasting Introduction Analysis of Chaotic Character Reconstruction of Phase Space Reconstruction of Phase Space The SVM Model [7-9] Immunogenetic Particle Swarm Incorporated Intelligence Algorithm Application and Analysis Choosing Samples and Chaos Analysis Prediction Process Predicted Values and Evaluating Indicator Conclusion References A Neural Model for Stereo Transparency with the Population of the Disparity Energy Models Introduction Neural Model for Stereo Transparency Definition of the Disparity Energy Model Population Activity to Stereo Images Reading Population Codes Performance of Disparity Detection Single Surface Transparent Surface Conclusion A Neuropsychologically-Inspired Computational Approach to the Generalization of Cerebellar Learning Introduction CMAC Network and the Cerebellar Learning Process Learning of Untrained CMAC Cells Case Study: Modeling of Human Glucose Metabolism Materials and Method Results Conclusions Introduction Related Works  Updated v-Support Vector Machines Conclusion References A New Mechanism on Brain Information Processing—Energy Coding Introduction Physical Model Model Formulation and Numerical Simulation Conclusion References A New Method for Multiple Spike Train Analysis Based on Information Discrepancy Introduction Method Experiment Symbolization of Spike Trains Function of Degree of Disagreement of the Symbolized Spike Trains Results Discussions A New SOM Algorithm for Electricity Load Forecasting Introduction The Self Organizing Maps A New Time Series Segmentation Algorithm Based on the SOM Experimental Results Conclusions and Future Research Trends A Novel Sequential Minimal Optimization Algorithm for Support Vector Regression Introduction Support Vector Regression SVR Dual Problem and Its Compact Form Optimality Condition for Problem 2 Global Convergence of SMO Algorithm for Solving Problem 2 Formulation of Subproblems Global Convergence of SMO Algorithm Proposed SMO Algorithm Working Set Selection Efficient Implementation of Algorithm 2 Simulation Results Conclusion A Novel Speech Processing Algorithm for Cochlear Implant Based on Selective Fundamental Frequency Control Introduction Algorithm Pre-process and Envelope Extraction in One Pathway Fundamental Frequency Extraction in the Other Pathway The Control Principle of Frequency Bands Selection Acoustic Simulation Experiment Results Discussion References A RBFN Based Cogging Force Estimator of PMLSM with q