NEURAL AND OPTICAL NEURAL NETWORKS

Authors
  1. Sala, K.L.
Corporate Authors
Communications Research Centre, Ottawa ONT (CAN)
Abstract
The report presents an overview of neural and optical neural network technologies which includes detailed descriptions of some of the more important paradigms such as single-and multi-layer perceptrons, backpropagation, Hopfield networks, associative memories, and stochastic learning. Optical neural networks are discussed from the point of view of two classes of implementations, namely, (a), those based upon the use of linear or nonlinear, two-dimensional optical elements such as spatial light modulators or bistable etalon arrays and, (b), those based upon the use of holographic elements. General conclusions regarding the present and future directions for neural and optical neural network research are presented along with specific recommendations for the conduct of an intramural/extramural research program into optical neural networks under the guidance of the Directorate of Optical Communications and Electrophotonics Technology of the Communications Research Center.
Report Number
CRC-1434 — Technical Report
Date of publication
15 Feb 1991
Number of Pages
77
DSTKIM No
91-01506
CANDIS No
68713
Format(s):
Hardcopy;Originator's fiche received by DSIS

Permanent link

Document 1 of 1

Date modified: