Abstract

Optically addressed ferroelectric liquid-crystal devices present certain characteristics that make them attractive for all-optical implementations of neural architectures. This is particularly true for the implementation of analog self-organizing systems such as Kohonen's self-organizing map. The optical architecture proposed here is based on the principle of a resonator memory. The main novelty lies in the two-dimensional, continuous, binary information encoding and organization. Basic operations such as neural decision, weight updating, and information spatial ordering are investigated. The principle of a learning procedure close to the Hebbian rule is discussed.

© 1994 Optical Society of America

All-optical implementation of a self-organizing map: Learning and taxonomy capability assessment

K. Heggarty, J. Duvillier, E. Carpio Pérez, and J. L. de Bougrenet de la Tocnaye
Appl. Opt. 34(35) 8167-8175 (1995)

Optoelectronic analogs of self-programming neural nets: architecture and methodologies for implementing fast stochastic learning by simulated annealing

Nabil H. Farhat
Appl. Opt. 26(23) 5093-5103 (1987)

64-channel correlator implementing a Kohonen-like neural network for handwritten-digit recognition

M. Barge, K. Heggarty, Y. Idan, and R. Chevallier
Appl. Opt. 35(23) 4655-4665 (1996)

Node-based reconfigurable volume interconnection. 1. Principles and optical design

B. Fracasso and J.-L. de Bougrenet de la Tocnaye
Appl. Opt. 33(23) 5348-5362 (1994)

High-capacity photorefractive neural network implementing a Kohonen topological map

Yann Frauel, Gilles Pauliat, André Villing, and Gérald Roosen
Appl. Opt. 40(29) 5162-5169 (2001)