Recent Advances in Artificial Neural Networks: (International Series on Computational Intelligence)

Neural networks represent a new generation of information processing paradigms designed to mimic-in a very limited sense-the human brain. They can learn, recall, and generalize from training data, and with their potential applications limited only by the imaginations of scientists and engineers, they are commanding tremendous popularity and research interest. Over the last four decades, researchers have reported a number of neural network paradigms, however, the newest of these have not appeared in book form-until now. Recent Advances in Artificial Neural Networks collects the latest neural network paradigms and reports on their promising new applications. World-renowned experts discuss the use of neural networks in pattern recognition, color induction, classification, cluster detection, and more. Application engineers, scientists, and research students from all disciplines with an interest in considering neural networks for solving real-world problems will find this collection useful.

Table of Contents:
A NEURO-SYMBOLIC HYBRID INTELLIGENT ARCHITECTURE WITH APPLICATIONS, J. Ghosh and I. Taha Knowledge Based Module for Representation of Initial Domain Knowledge Extraction of Supplementary Rules via the Statistical Analysis Module The Mapping Module The Discretization Module Refining Input Characterization Rule Extraction Rule Evaluation and Ordering Procedure for the Refined Expert System The Integrated Decision Maker Application: Controlling Water Reservoirs Application of the Statistical Approach Discussion NEW RADICAL BASIS NEURAL NETWORKS AND THEIR APPLICATION IN A LARGE-SCALE HANDWRITTEN DIGIT RECOGNITION PROBLEM, N.B. Karayiannis and S. Behnke Function Approximation Models and RBF Neural Networks Reformulating Radial Basis Neural Networks Admissible Generator Functions Selecting Generator Functions Learning Algorithms Based on Gradient Descent Generator Functions and Gradient Descent Learning Handwritten Digit Recognition Conclusions EFFICIENT NEURAL NETWORK-BASED METHODOLOGY FOR THE DESIGN OF MULTIPLE CLASSIFIERS, N. Vassilas Proposed Methodology Modifications of Supervised Algorithms Multimodular Classification Land-Cover Classification Summary LEARNING FINE MOTION IN ROBOTICS: DESIGN AND EXPERIMENTS, C. Versino and L.M. Gambardella How to Find the Path? The Model-Based System The Sensor-Based System Perception Clustering Action Triggering All Together Why Use a SOM-Like Network? Planner vs. HEKM Conclusions A NEW NEURAL NETWORK FOR ADAPTIVE PATTERN RECOGNITION OF MULTICHANNEL INPUT SIGNALS, M. Fernández-Delgado, J. Presedo, M. Lama, and S. Barro Architecture and Functionality of MART Learning in MART Analysis of the Behavior of Certain Adaptive Parameters A Real Application Example Discussion LATERAL PRIMING ADAPTIVE RESONANCE THEORY (LAPART)-2: INNOVATION IN ART, T.P. Caudell and M.J. Healy ART-1, Stacknet, and LAPART-1 The LAPART-2 Algorithm The Learning Theorems Numerical Experiments Discussion Conclusion NEURAL NETWORK LEARNING IN A TRAVEL RESERVATION DOMAIN, H.A. Aboulenien and P. De Wilde Agents Neural Network Role Agent Architecture Operation Summary RECENT ADVANCES IN NEURAL NETWORK APPLICATIONS IN PROCESS CONTROL, U. Halici, K. Leblebicioglu, C. Özgen, and S. Tuncay Process Control Use of Neural Networks in Control Case Study I: pH Control in Neutralization System Case Study II: Adaptive Nonlinear-Model Predictive Control Using Neural Networks for Control of High Purity Industrial Distillation Column Case Study III: PI Controller for a Batch Distillation Column with Neural Network Coefficient Estimator Case Study IV: A Rule-Based Neuro-Optimal Controller for Steam-Jacketed Kettle Remarks and Future Studies MONITORING INTERNAL COMBUSTION ENGINES BY NEURAL NETWORK BASED VIRTUAL SENSING, R.J. Howlett, M.M. de Zoysa, and S.D. Walters The Engine Management system Virtual Sensor Systems Air-Fuel Ratio Combustion Monitoring Using the Spark Plug The Ignition System of a Spark-Ignition Engine Neural-Networks for Use in Virtual Sensors AFR Estimation Using Neural Network Spark Voltage Characterization Conclusions NEURAL ARCHITECTURES OF FUZZY PETRI NETS, W. Pedrycz Introduction The Generalization of the Petri Net and its Underlying Architecture The Architecture of the Fuzzy Petri Net The Learning Procedure Interfacing Fuzzy Petri Nets with Granular Information Experiments Conclusions INDEX Each chapter includes an introduction and references.