Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Hopfield neural network using electron-trapping materials

Open Access Open Access

Abstract

The electron-trapping (ET) materials are optical storage materials developed by Quantex. When ET materials are exposed to visible light (e.g., 488 nm), electrons from the ground state are excited and then captured by thermally stable traps. Subsequent exposure to IR light (e.g., 1064 nm) excites and releases the trapped electrons to the ground state with the emission of orange-to-red light. The ET materials are capable of performing multiplication, addition, and subtraction within a dynamic range covering four orders of magnitude. The orange emission intensity is proportional to the product of the blue write-in intensity and the IR read-out intensity. The addition and subtraction operations are performed by increasing and decreasing the number of trapped electrons. An ET thin film can be used as an optical mask to perform matrix vector multiplication. Furthermore, the ET thin film has an inherent capability for summing, subtracting, and storing optical data, in addition to multiplication. This extra capability, which is not commonly found in spatial light modulators, will be used for the optical formation of an interconnection matrix (prescribed learning).

© 1990 Optical Society of America

PDF Article
More Like This
Three-dimensional stacked-layer optical memory using electron-trapping materials

Suganda Jutamulia, George M. Storti, William Seiderman, and Joseph Lindmayer
ThT5 OSA Annual Meeting (FIO) 1990

Optical neural network using novel electron trapping materials

Francis T. S. Yu, Xiang Y. Yang, Shudong Wu, Suganda Jutamulia, George M. Storti, and Joseph Lindmayer
MV3 OSA Annual Meeting (FIO) 1991

Novel electron trapping (ET) spatial light modulator

Suganda Jutamulia, George M. Storti, Joseph Lindmayer, and William Seiderman
TuC2 Spatial Light Modulators and Applications (SLM) 1990

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.