Optimal Unsupervised Learning in Feedforward Neural Networks

Title: en_US Optimal Unsupervised Learning in Feedforward Neural Networks
Creator: en_US Sanger, Terence D.
Date: 2004-10-20T20:11:57Z
Date Available: 2004-10-20T20:11:57Z
Date Issued: en_US 1989-01-01
Identifier: en_US AITR-1086
uri: http://hdl.handle.net/1721.1/6976
Abstract: en_US We investigate the properties of feedforward neural networks trained with Hebbian learning algorithms. A new unsupervised algorithm is proposed which produces statistically uncorrelated outputs. The algorithm causes the weights of the network to converge to the eigenvectors of the input correlation with largest eigenvalues. The algorithm is closely related to the technique of Self-supervised Backpropagation, as well as other algorithms for unsupervised learning. Applications of the algorithm to texture processing, image coding, and stereo depth edge detection are given. We show that the algorithm can lead to the development of filters qualitatively similar to those found in primate visual cortex.
Extent: 8663770 bytes; 6747778 bytes
Format: application/postscript; application/pdf
Language: en_US
Relation: en_US AITR-1086