A real-time learning algorithm for a multilayered neural network based on the extended Kalman filter

Author

Iiguni, Youji ; Sakai, Hideaki ; Tokumaru, Hidekatsu

Author_Institution

Fac. of Eng., Kyoto Univ., Japan

Volume

40

Issue

4

fYear

1992

fDate

4/1/1992 12:00:00 AM

Firstpage

959

Lastpage

966

Abstract

A novel real-time learning algorithm for a multilayered neural network is derived from the extended Kalman filter (EKF). Since this EKF-based learning algorithm approximately gives the minimum variance estimate of the linkweights, the convergence performance is improved in comparison with the backwards error propagation algorithm using the steepest descent techniques. Furthermore, tuning parameters which crucially govern the convergence properties are not included, which makes its application easier. Simulation results for the XOR and parity problems are provided

Keywords

Kalman filters; learning systems; neural nets; XOR problems; convergence performance; extended Kalman filter; minimum variance estimate; multilayered neural network; parity problems; real-time learning algorithm; simulation results; Application software; Backpropagation algorithms; Convergence; Iterative algorithms; Linear systems; Multi-layer neural network; Neural networks; Nonlinear systems; Parameter estimation; Signal processing algorithms;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.127966

Filename

127966