Abstract

Optical implementation of a backpropagating neuron by means of a nonlinear Fabry-Perot etalon requires thresholding a forward signal beam while the transmittance of a backpropagating beam is multiplied by the differential of the forward signal. This is achievable by inputting a bichromatic field to a three-level system in an optical cavity. The response characteristics of this device have the added possibility of adaptability of the threshold by the backward probe input intensity.

© 1989 Optical Society of America

Multilayer optical learning networks

Kelvin Wagner and Demetri Psaltis
Appl. Opt. 26(23) 5061-5076 (1987)

Infrared nonlinear neurons using the field shielding effect in CdTe

Mehrdad Ziari, William H. Steier, and Robert L. S. Devine
Appl. Opt. 29(14) 2074-2083 (1990)

Backpropagation through nonlinear units for the all-optical training of neural networks

Xianxin Guo, Thomas D. Barrett, Zhiming M. Wang, and A. I. Lvovsky
Photon. Res. 9(3) B71-B80 (2021)

Subtracting incoherent optical neuron model: analysis, experiment, and applications

Chein-Hsun Wang and B. Keith Jenkins
Appl. Opt. 29(14) 2171-2186 (1990)

Adaptive neural network in a hybrid optical/electronic architecture using lateral inhibition

Peter J. de Groot and Robert J. Noll
Appl. Opt. 28(18) 3852-3859 (1989)