Building Neural Networks / Edition 1

by David M. Skapura

Overview

Organized by application areas, rather than by specific network architectures or learning algorithms, Builiding Neural Networks shows why certain networks are more suitable than others for solving specific kinds of problems. Skapura also reviews principles of neural information processing and furnishes an operations summary of the most popular neural-network processing models. Finally, the book provides information on the practical aspects of application design, and contains six topic-oriented chapters on specific applications of neural-network systems. These applications include networks that perform:

• Pattern matching, storage, and recall
• Business and financial systems
• Data extraction from images
• Mechanical process control systems
• New neural networks that combine pattern matching with fuzzy logic

The book includes application-oriented excercises that further help you see how a neural network solves a problem, and that reinforce your understanding of modeling techniques.

Product Details

ISBN-13:	2900201539218
Publisher:	Addison-Wesley
Publication date:	12/05/1995
Series:	ACM Press Series
Edition description:	New Edition
Pages:	286
Product dimensions:	6.00(w) x 1.25(h) x 9.00(d)

About the Author

About David M. Skapura

David M. Skapura is the coauthor, with James A. Freeman, of Neural Networks: Algorithms, Applications, and Programming Techniques. He is currently employed by Brightware Corporation (a spin-off of Inference Corporation), where he works as an applications consultant, developing customized knowledge-based systems and applications. He is also the founder and president of Scient Computing, a small Houston consulting firm specializing in neural-networking applications and research. Previously at Loral Space Information Systems, Skapura investigated the applicability of neural networks to NASA's advanced automation requirements. He is an adjunct professor at the University of Houston at Clear Lake.

Table of Contents

Foundations.

Motivation.
Neural-Network Fundamentals.
Single Neuron Computations.
Network Computations.
Network Simulation.
Foundations Summary.
Suggested Readings.
Bibliography.

Paradigms.

The Backpropagation Network.
The Counterpropagation Network.
Adaptive Resonance Theory.
The Multidirectional Associative Memory.
The Hopfield Memory.
Network-Learning Summary.
Suggested Readings.
Bibliography.

Application Design.

Developing a Data Representation.
Pattern Representation Methods.
Exemplar Analysis.
Training and Performance Evaluation.
A Practical Example.
Application-Design Summary.
Suggested Readings.
Bibliography.

Associative Memories.

Associative-Memory Definitions.
Character Recognition.
State-Space Search.
Image Interpolation.
Diagnostic Aids.
Associative-Memory Summary.
Suggested Readings.
Bibliography.

Business and Financial Applications.

Financial Modeling.
Market Prediction.
Bond Rating.
Predicting Commodity Futures.
Financial-Applications Summary.
Suggested Readings.
Bibliography.

Pattern Classification.

NETtalk.
Radar-Signature Classifier.
Prostate-Cancer Detection.
Pattern-Classification Summary.
Suggested Readings.
Bibliography.

Image Processing.

Image-Processing Networks.
Gender Recognition from Facial Images.
Imagery Feature Discovery.
Aircraft Tracking in Video Imagery.
Image-Processing Summary.
Suggested Readings.
Bibliography.

Process Control and Robotics.

Control Theory.
Cart/Pole Balancer.
Bipedal-Locomotion Control.
Robotic Manipulator Control.
Control-Application Summary.
Suggested Readings.
Bibliography.

Fuzzy Neural Systems.

Fuzzy Logic.
Implementation of a Fuzzy Network.
Fuzzy Neural Inference.
Fuzzy Control of BPN Learning.
Fuzzy Neural-System Summary.
Suggested Readings.
Bibliography.

Answers to Selected Exercises.
Index.

Editorial Reviews

A practical introduction to the problems neural network technology can solve. Intended for upper-level undergraduate or first-year graduate students in engineering or computer science courses or for industry engineers who wish to explore practical aspects of neural networking. Requires a good understanding of vector and matrix mathematics. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews