A hybrid global learning algorithm based on global search and least squares techniques for backpropagation networks

Author

Leung, Chi-Tat ; Chow, Tommy W S

Author_Institution

Dept. of Electron. Eng., City Univ. of Hong Kong, Kowloon, Hong Kong

Volume

3

fYear

1997

fDate

9-12 Jun 1997

Firstpage

1890

Abstract

A hybrid learning algorithm for backpropagation network based on global search and least squares methods is presented to increase the speed of convergence. The proposed algorithm comprises global search and least squares parts. The global search part trains a backpropagation network over a reduced weight space. The remained weights are calculated in accordance with linear least squares method. Two problems of nonlinear function approximation and modified XOR are applied to demonstrate the fast global search performance of the proposed algorithm. The results indicate that the proposed algorithm enables the learning process to significantly speed up by at most 4670% in terms of iterations and do not trap in local minima

Keywords

backpropagation; convergence; least squares approximations; neural nets; search problems; EXOR; backpropagation networks; global search techniques; hybrid global learning algorithm; linear least-squares method; modified XOR; nonlinear function approximation; reduced weight space; Approximation algorithms; Backpropagation algorithms; Convergence; Function approximation; Least squares methods; Neurons; Nonlinear systems; Robots; Search methods; Supervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks,1997., International Conference on

Conference_Location

Houston, TX

Print_ISBN

0-7803-4122-8

Type

conf

DOI

10.1109/ICNN.1997.614187

Filename

614187