Using the Karhunen-Loe´ve transformation in the back-propagation training algorithm

Author

Malki, H.A. ; Moghaddamjoo, A.

Author_Institution

Dept. of Electr. Eng. & Comput. Sci., Wisconsin Univ., Milwaukee, WI, USA

Volume

2

Issue

1

fYear

1991

fDate

1/1/1991 12:00:00 AM

Firstpage

162

Lastpage

165

Abstract

A novel training approach based on the back-propagation algorithm is introduced. In the proposed approach, initially, a set of training vectors is obtained by applying the Karhunen-Loe´ve transform on the training patterns. The training is first started in the direction of the major eigenvectors of the correlation matrix of the training patterns and then continues by gradually including the remaining components, in their order of significance. With this approach, the number of computations is significantly reduced and the learning rate is improved. The performance of this method is compared with the standard back-propagation algorithm in segmenting a synthetic noisy image

Keywords

eigenvalues and eigenfunctions; learning systems; matrix algebra; neural nets; pattern recognition; transforms; Karhunen-Loe´ve transformation; back-propagation training algorithm; correlation matrix; major eigenvectors; pattern recognition; segmentation; synthetic noisy image; Associative memory; Circuits; Computer networks; Convergence; Dispersion; Neural networks; Optical feedback; Optical propagation; Optimized production technology; Physics computing;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/72.80306

Filename

80306