A learning algorithm for recurrent neural networks and its application to nonlinear identification

Author

Yamamoto, Yoshihiro ; Nikiforuk, Peter N.

Author_Institution

Tottori Univ., Japan

fYear

1999

fDate

1999

Firstpage

551

Lastpage

556

Abstract

A new learning algorithm is presented for a supervised learning of recurrent neural networks without using a gradient method. First, fictitious teacher signals for the outputs of each hidden unit are algebraically determined by an error backpropagation (EBP) method. Then, the weight parameters are determined by using an exponentially weighted least squares (EWLS) method. This is called EBP-EWLS algorithm which is an extension of the algorithm for a multilayer neural network. The algorithm is applied for identification of a nonlinear system to show the effectiveness of the proposed method and a new idea for nonlinear identification

Keywords

backpropagation; gradient methods; identification; least squares approximations; nonlinear systems; recurrent neural nets; error backpropagation; exponentially weighted least squares; gradient method; identification; nonlinear system; recurrent neural networks; supervised learning; Convergence; Ear; Gradient methods; Least squares methods; Multi-layer neural network; Neural networks; Nonlinear systems; Recurrent neural networks; Supervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Aided Control System Design, 1999. Proceedings of the 1999 IEEE International Symposium on

Conference_Location

Kohala Coast, HI

Print_ISBN

0-7803-5500-8

Type

conf

DOI

10.1109/CACSD.1999.808707

Filename

808707