Abstract

Adaptive optical systems such as the bipolar, polarization-based optical connectionist machine are capable of operating in the presence of substantial noise generated by optoelectronic devices such as spatial light modulators, sources, and detectors. We present results on two optoelectronic connectionist machines that implement the single-layer delta rule and backward error propagation neural network algorithms and analyze the influence of noise on their performance. Results show that an optoelectronic neural network with 200 input units can easily classify 30 random patterns with a spatial light modulator contrast ratio of 10:1 and output cross talk of 10%.

© 1992 Optical Society of America

Motivations for using ferroelectric liquid crystal spatial light modulators in neurocomputing

Kristina M. Johnson and Garret Moddel
Appl. Opt. 28(22) 4888-4899 (1989)

Optoelectronic array that computes error and weight modification for a bipolar optical neural network

C. C. Mao and K. M. Johnson
Appl. Opt. 32(8) 1290-1296 (1993)

Design and analysis of fixed planar holographic interconnects for optical neural networks

Paul E. Keller and Arthur F. Gmitro
Appl. Opt. 31(26) 5517-5526 (1992)

Four-quadrant optical matrix–vector multiplication machine as a neural-network processor

S. Abramson, D. Saad, E. Marom, and N. Konforti
Appl. Opt. 32(8) 1330-1337 (1993)

Hybrid optoelectronic adaptive resonance theory neural processor, ART1

Thomas P. Caudell
Appl. Opt. 31(29) 6220-6229 (1992)