A Study of Self-Organization Method of Neural Networks

Shigenobu Yamawaki

doi:10.11499/sicep.2002.0.352.0

SICE Annual Conference 2002

DOI https://doi.org/10.11499/sicep.2002.0.352.0

Conference information

Host: The Society of Instrument and Control Engineers

Co-host: IEEE/Industiral Electronic Society, IEEE/Robotics and Automation Society, IEEE/Control System Society

A Study of Self-Organization Method of Neural Networks

Shigenobu Yamawaki

Author information

Keywords: Forgetting, Self-Organization, Regression Neural Network, Least Squares Method, Back propagation, Identification

CONFERENCE PROCEEDINGS FREE ACCESS

Pages 352

Details

Abstract

The ability to describe a neural network depends greatly on the number of neurons. In this paper, we consider a method of obtaining an optimal neural network through a forgetting operation. It is shown that by using the proposed method, an optimum neural network can be constructed not only from neurons having large singular values of the coefficient matrix, but also from neurons having small singular values of the coefficient matrix.

Corresponding author

Register with J-STAGE for free!