Abstract
A novel type of quantized learning rule with unipolar binary weights that is useful for the optical implementation of neural networks is reported. An input-dependent thresholding operation is also proposed to remove the unwanted effects that are due to the insufficient contrast ratio of spatial light modulators as synaptic connection devices. Moreover, we experimentally demonstrate the recognition of 26 characters of the alphabet by using the single set of an optoelectronic three-layered network.
© 1990 Optical Society of America
Full Article | PDF ArticleMore Like This
A. Von Lehmen, E. G. Paek, P. F. Liao, A. Marrakchi, and J. S. Patel
Opt. Lett. 14(17) 928-930 (1989)
Fumihiko Itoh, Ken-ichi Kitayama, and Yasuaki Tamura
Opt. Lett. 15(15) 860-862 (1990)
Masaya Oita, Jun Ohta, Shuichi Tai, and Kazuo Kyuma
Opt. Lett. 15(4) 227-229 (1990)