Title :
2D sparse distributed memory-optical neural network for pattern recognition
Author :
Mayol-Cuevas, Walterio W. ; Gómez-Ramìrez, Eduardo
Author_Institution :
Lab. de Sistemas Complejos, Mexico
fDate :
27 Jun-2 Jul 1994
Abstract :
An implementation of a Kanerva neural network using optical techniques is presented. The advantages of this paradigm over conventional algorithms in 2D signal recognition and reconstruction tasks are shown, as its ease of implementation using optical methods in the estimation of the weight matrix, reducing learning time
Keywords :
distributed memory systems; image recognition; image reconstruction; optical neural nets; optical storage; 2D signal recognition; 2D sparse distributed memory-optical neural network; Kanerva neural network; learning time; pattern recognition; signal reconstruction; weight matrix estimation; Associative memory; Hamming distance; Laboratories; Neural networks; Optical arrays; Optical computing; Optical fiber networks; Pattern classification; Pattern recognition; Signal reconstruction;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374541
