Artificial Neural Networks: Formal Models and Their Applications – ICANN 2005 15th International Conference, Warsaw, Poland, September 11-15, 2005, Proceedings, Part II

This volume is the first part of the two-volume proceedings of the International C- ference on Artificial Neural Networks (ICANN 2005), held on September 11–15, 2005 in Warsaw, Poland, with several accompanying workshops held on September 15, 2005 at the Nicolaus Copernicus University, Toru , Poland. The ICANN conference is an annual meeting organized by the European Neural Network Society in cooperation with the International Neural Network Society, the Japanese Neural Network Society, and the IEEE Computational Intelligence Society. It is the premier European event covering all topics concerned with neural networks and related areas. The ICANN series of conferences was initiated in 1991 and soon became the major European gathering for experts in those fields. In 2005 the ICANN conference was organized by the Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland, and the Nicolaus Copernicus Univ- sity, Toru , Poland. From over 600 papers submitted to the regular sessions and some 10 special c- ference sessions, the International Program Committee selected – after a thorough peer-review process – about 270 papers for publication. The large number of papers accepted is certainly a proof of the vitality and attractiveness of the field of artificial neural networks, but it also shows a strong interest in the ICANN conferences.

New Neural Network Models.- Neuro-fuzzy Kolmogorov’s Network.- A Neural Network Model for Inter-problem Adaptive Online Time Allocation.- Discriminant Parallel Perceptrons.- A Way to Aggregate Multilayer Neural Networks.- Generalized Net Models of MLNN Learning Algorithms.- Monotonic Multi-layer Perceptron Networks as Universal Approximators.- Short Term Memories and Forcing the Re-use of Knowledge for Generalization.- Interpolation Mechanism of Functional Networks.- Supervised Learning Algorithms.- Neural Network Topology Optimization.- Rough Sets-Based Recursive Learning Algorithm for Radial Basis Function Networks.- Support Vector Neural Training.- Evolutionary Algorithms for Real-Time Artificial Neural Network Training.- Developing Measurement Selection Strategy for Neural Network Models.- Nonlinear Regression with Piecewise Affine Models Based on RBFN.- Batch-Sequential Algorithm for Neural Networks Trained with Entropic Criteria.- Multiresponse Sparse Regression with Application to Multidimensional Scaling.- Training Neural Networks Using Taguchi Methods: Overcoming Interaction Problems.- A Global-Local Artificial Neural Network with Application to Wave Overtopping Prediction.- Ensemble-Based Learning.- Learning with Ensemble of Linear Perceptrons.- Combination Methods for Ensembles of RBFs.- Ensemble Techniques for Credibility Estimation of GAME Models.- Combination Methods for Ensembles of MF.- New Results on Ensembles of Multilayer Feedforward.- Unsupervised Learning.- On Variations of Power Iteration.- Linear Dimension Reduction Based on the Fourth-Order Cumulant Tensor.- On Spectral Basis Selection for Single Channel Polyphonic Music Separation.- Independent Subspace Analysis Using k-Nearest Neighborhood Distances.- Recurrent Neural Networks.- Study of the Behavior of a New Boosting Algorithm for Recurrent Neural Networks.- Time Delay Learning by Gradient Descent in Recurrent Neural Networks.- Representation and Identification Method of Finite State Automata by Recurrent High-Order Neural Networks.- Global Stability Conditions of Locally Recurrent Neural Networks.- Reinforcement Learning.- An Agent-Based PLA for the Cascade Correlation Learning Architecture.- Dual Memory Model for Using Pre-existing Knowledge in Reinforcement Learning Tasks.- Stochastic Processes for Return Maximization in Reinforcement Learning.- Maximizing the Ratio of Information to Its Cost in Information Theoretic Competitive Learning.- Completely Self-referential Optimal Reinforcement Learners.- Model Selection Under Covariate Shift.- Bayesian Approaches to Learning.- Smooth Performance Landscapes of the Variational Bayesian Approach.- Jacobi Alternative to Bayesian Evidence Maximization in Diffusion Filtering.- Bayesian Learning of Neural Networks Adapted to Changes of Prior Probabilities.- A New Method of Learning Bayesian Networks Structures from Incomplete Data.- Bayesian Hierarchical Ordinal Regression.- Traffic Flow Forecasting Using a Spatio-temporal Bayesian Network Predictor.- Learning Theory.- Manifold Constrained Variational Mixtures.- Handwritten Digit Recognition with Nonlinear Fisher Discriminant Analysis.- Separable Data Aggregation in Hierarchical Networks of Formal Neurons.- Induced Weights Artificial Neural Network.- SoftDoubleMinOver: A Simple Procedure for Maximum Margin Classification.- On the Explicit Use of Example Weights in the Construction of Classifiers.- A First Approach to Solve Classification Problems Based on Functional Networks.- A Description of a Simulation Environment and Neural Architecture for A-Life.- Neural Network Classifers in Arrears Management.- Sequential Classification of Probabilistic Independent Feature Vectors Based on Multilayer Perceptron.- Multi-class Pattern Classification Based on a Probabilistic Model of Combining Binary Classifiers.- Evaluating Performance of Random Subspace Classifier on ELENA Classification Database.- Artificial Neural Networks for System Modeling, Decision Making, Optimalization and Control.- A New RBF Neural Network Based Non-linear Self-tuning Pole-Zero Placement Controller.- Using the Levenberg-Marquardt for On-line Training of a Variant System.- Optimal Control Yields Power Law Behavior.- A NeuroFuzzy Controller for 3D Virtual Centered Navigation in Medical Images of Tubular Structures.- Emulating Process Simulators with Learning Systems.- Evolving Modular Fast-Weight Networks for Control.- Topological Derivative and Training Neural Networks for Inverse Problems.- Application of Domain Neural Network to Optimization Tasks.- Eigenvalue Problem Approach to Discrete Minimization.- A Neurocomputational Approach to Decision Making and Aging.- Comparison of Neural Network Robot Models with Not Inverted and Inverted Inertia Matrix.- Causal Neural Control of a Latching Ocean Wave Point Absorber.- An Off-Policy Natural Policy Gradient Method for a Partial Observable Markov Decision Process.- A Simplified Forward-Propagation Learning Rule Applied to Adaptive Closed-Loop Control.- Improved, Simpler Neural Controllers for Lamprey Swimming.- Supervision of Control Valves in Flotation Circuits Based on Artificial Neural Network.- Comparison of Volterra Models Extracted from a Neural Network for Nonlinear Systems Modeling.- Identification of Frequency-Domain Volterra Model Using Neural Networks.- Hierarchical Clustering for Efficient Memory Allocation in CMAC Neural Network.- Special Session: Knowledge Extraction from Neural Networks Organizer and Chair: D. A. Elizondo.- Knowledge Extraction from Unsupervised Multi-topographic Neural Network Models.- Current Trends on Knowledge Extraction and Neural Networks.- Prediction of Yeast Protein–Protein Interactions by Neural Feature Association Rule.- A Novel Method for Extracting Knowledge from Neural Networks with Evolving SQL Queries.- CrySSMEx, a Novel Rule Extractor for Recurrent Neural Networks: Overview and Case Study.- Computational Neurogenetic Modeling: Integration of Spiking Neural Networks, Gene Networks, and Signal Processing Techniques.- Information Visualization for Knowledge Extraction in Neural Networks.- Combining GAs and RBF Neural Networks for Fuzzy Rule Extraction from Numerical Data.- Temporal Data Analysis, Prediction and Forecasting.- Neural Network Algorithm for Events Forecasting and Its Application to Space Physics Data.- Counterpropagation with Delays with Applications in Time Series Prediction.- Bispectrum-Based Statistical Tests for VAD.- Back-Propagation as Reinforcement in Prediction Tasks.- Mutual Information and k-Nearest Neighbors Approximator for Time Series Prediction.- Some Issues About the Generalization of Neural Networks for Time Series Prediction.- Multi-step-ahead Prediction Based on B-Spline Interpolation and Adaptive Time-Delay Neural Network.- Support Vector Machines and Kernel-Based Methods.- Training of Support Vector Machines with Mahalanobis Kernels.- Smooth Bayesian Kernel Machines.- A New Kernel-Based Algorithm for Online Clustering.- The LCCP for Optimizing Kernel Parameters for SVM.- The GCS Kernel for SVM-Based Image Recognition.- Informational Energy Kernel for LVQ.- Reducing the Effect of Out-Voting Problem in Ensemble Based Incremental Support Vector Machines.- A Comparison of Different Initialization Strategies to Reduce the Training Time of Support Vector Machines.- A Hierarchical Support Vector Machine Based Solution for Off-line Inverse Modeling in Intelligent Robotics Applications.- LS-SVM Hyperparameter Selection with a Nonparametric Noise Estimator.- Building Smooth Neighbourhood Kernels via Functional Data Analysis.- Recognition of Heartbeats Using Support Vector Machine Networks – A Comparative Study.- Componentwise Support Vector Machines for Structure Detection.- Memory in Backpropagation-Decorrelation O(N) Efficient Online Recurrent Learning.- Soft Computing Methods for Data Representation, Analysis and Processing.- Incremental Rule Pruning for Fuzzy ARTMAP Neural Network.- An Inductive Lea