Share Email Print

Proceedings Paper

Renewal method of weight matrix in optical neural network
Author(s): Ichiro Tohyama; Yoshio Hayasaki; Toyohiko Yatagai; Masahiko Mori; Satoshi Ishihara
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

We proposed a new renewal method of the weights to realize a back propagation learning algorithm with a large number of iteration times in an optical neural network. The main purpose of this method is to use the limited dynamic range of an optical device widely. Since some weights displaying on the device are quickly saturated in the conventional method, the learning is not completed in many cases. The new technique can suppress saturation of the weights so that learning is completed. The computer simulation and the optical experiment are presented.

Paper Details

Date Published: 12 July 1993
PDF: 8 pages
Proc. SPIE 1806, Optical Computing, (12 July 1993); doi: 10.1117/12.147840
Show Author Affiliations
Ichiro Tohyama, Univ. of Tsukuba (Japan)
Yoshio Hayasaki, Univ. of Tsukuba (Japan)
Toyohiko Yatagai, Univ. of Tsukuba (Japan)
Masahiko Mori, Electrotechnical Lab. (Japan)
Satoshi Ishihara, Electrotechnical Lab. (Japan)

Published in SPIE Proceedings Vol. 1806:
Optical Computing
Andrey M. Goncharenko; Fedor V. Karpushko; George V. Sinitsyn; Sergey P. Apanasevich, Editor(s)

© SPIE. Terms of Use
Back to Top