Abstract
A gain and exposure schedule that theoretically eliminates the effect of photorefractive weight decay for the general class of outer-product neural-network learning algorithms (e.g., backpropagation, Widrow – Hoff, perceptron) is presented. This schedule compensates for photorefractive diffraction-efficiency decay by iteratively increasing the spatial-light-modulator transfer function gain and decreasing the weight-update exposure time. Simulation results for the scheduling procedure, as applied to backpropagation learning for the exclusive-or problem, show improved learning performance compared with results for networks trained without scheduling.
© 1995 Optical Society of America
Full Article | PDF ArticleMore Like This
Chau-Jern Cheng, Pochi Yeh, and Ken Yuh Hsu
J. Opt. Soc. Am. B 11(9) 1619-1624 (1994)
Gregory C. Petrisor, Adam A. Goldstein, B. Keith Jenkins, Edward J. Herbulock, and Armand R. Tanguay
Appl. Opt. 35(8) 1328-1343 (1996)
T. J. Allen, K. M. Curtis, and J. W. Orton
Appl. Opt. 34(20) 4136-4139 (1995)