Principal Component Neural Networks: Theory and Applications / Edition 1

by K. I. Diamantaras, S. Y. Kung

Overview

Systematically explores the relationship between principal component analysis (PCA) and neural networks. Provides a synergistic examination of the mathematical, algorithmic, application and architectural aspects of principal component neural networks. Using a unified formulation, the authors present neural models performing PCA from the Hebbian learning rule and those which use least squares learning rules such as back-propagation. Examines the principles of biological perceptual systems to explain how the brain works. Every chapter contains a selected list of applications examples from diverse areas.

Product Details

ISBN-13:	9780471054368
Publisher:	Wiley
Publication date:	03/08/1996
Series:	Adaptive and Cognitive Dynamic Systems: Signal Processing, Learning, Communications and Control , #4
Pages:	272
Product dimensions:	6.34(w) x 9.49(h) x 0.79(d)

About the Author

K. I. Diamantaras is a research scientist at Aristotle University in Thessaloniki, Greece. He received his PhD from Princeton University and was formerly a research scientist for Siemans Corporate Research.

S. Y. Kung is Professor of Electrical Engineering at Princeton University and received his PhD from Stanford University. He was formerly a professor of electrical engineering at the University of Southern California.

Table of Contents

A Review of Linear Algebra.

Principal Component Analysis.

PCA Neural Networks.

Channel Noise and Hidden Units.

Heteroassociative Models.

Signal Enhancement Against Noise.

VLSI Implementation.

Appendices.

Bibliography.

Index.

Editorial Reviews

Explicates the relationship between principle component analysis (PCA) and neural networks, studying issues pertaining to both neural network models, such as network structures and algorithms, and theoretical extensions/generalizations of PCA. Includes reference mathematical appendices. Of interest to readers with background in college calculus and probability theory, in disciplines such as mathematics; statistics; neuropsychology; artificial intelligence; and engineering. Annotation c. Book News, Inc., Portland, OR (booknews.com)

Booknews