Abstract
The effect of discretizing interconnection weight strengths in an optoelectronic learning neural network based on the backpropagation algorithm is investigated. We discuss how discretization arises in such an implementation. Using computer simulations we find that learning performance, as tested on the two-input XOR problem, is poor but that the addition of noise to the system results in substantial improvement.
© 1989 Optical Society of America
Full Article | PDF ArticleMore Like This
Masaya Oita, Masanobu Takahashi, Shuichi Tai, and Kazuo Kyuma
Opt. Lett. 15(21) 1227-1229 (1990)
T. J. Allen, K. M. Curtis, and J. W. Orton
Appl. Opt. 34(20) 4136-4139 (1995)
Adam A. Goldstein, Gregory C. Petrisor, and B. Keith Jenkins
Opt. Lett. 20(6) 611-613 (1995)