Share Email Print

Proceedings Paper

Efficient optical architecture for sparsely connected neural networks
Author(s): Butler P. Hine; John D. Downie; Max B. Reid
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

An architecture for general-purpose optical neural network processor is presented in which the interconnections and weights are formed by directing coherent beams holographically, thereby making use of the space-bandwidth products of the recording medium for sparsely interconnected networks more efficiently that the commonly used vector-matrix multiplier, since all of the hologram area is in use. An investigation is made of the use of computer-generated holograms recorded on such updatable media as thermoplastic materials, in order to define the interconnections and weights of a neural network processor; attention is given to limits on interconnection densities, diffraction efficiencies, and weighing accuracies possible with such an updatable thin film holographic device.

Paper Details

Date Published: 1 September 1990
PDF: 11 pages
Proc. SPIE 1296, Advances in Optical Information Processing IV, (1 September 1990); doi: 10.1117/12.21283
Show Author Affiliations
Butler P. Hine, NASA/Ames Research Ctr. (United States)
John D. Downie, NASA/Ames Research Ctr. (United States)
Max B. Reid, NASA/Ames Research Ctr. (United States)

Published in SPIE Proceedings Vol. 1296:
Advances in Optical Information Processing IV
Dennis R. Pape, Editor(s)

© SPIE. Terms of Use
Back to Top