The eigenspace separation transform for neural-network classifiers

This paper presents a linear transform that compresses data in a manner des igned to improve the performance of a neural network used as a binary class ifier. The classifier is intended to accommodate data distributions that ma y be non-normal, may have equal class means, may be multimodal, and have un known a priori probabilities for the two classes. The transform, which is c alled the eigenspace separation transform, allows the reduction of the size of a neural network while enhancing its generalization accuracy as a binar y classifier. Published by Elsevier Science Ltd.