Abstract
A reconfigurable optical crossbar switch has been integrated into a microcomputer environment for modeling neural networks that learn. A differential technique is used for obtaining bipolar inputs, outputs, and connection weights in a mask-type crossbar using an inexpensive low-contrast spatial light modulator. Neural networks based on matched-filter techniques and competitive interconnects have been implemented for adaptive associative recall as well as unsupervised learning in an autonomous controller.
© 1988 Optical Society of America
PDF ArticleMore Like This
Shudong Wu, Taiwei Lu, and Francis T. S. Yu
THHH2 OSA Annual Meeting (FIO) 1988
KELVIN WAGNER
WK2 Conference on Lasers and Electro-Optics (CLEO:S&I) 1988
G. C. Petrisor, B. K. Jenkins, H. Chin, and A. R. Tanguay
MVV1 OSA Annual Meeting (FIO) 1990