Qiao, Yong and Psaltis, Demetri (1992) Local learning algorithm for optical neural networks. Applied Optics, 31 (17). pp. 3285-3288. ISSN 0003-6935 http://resolver.caltech.edu/CaltechAUTHORS:QIAao92
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:QIAao92
An anti-Hebbian local learning algorithm for two-layer optical neural networks is introduced. With this learning rule, the weight update for a certain connection depends only on the input and output of that connection and a global, scalar error signal. Therefore the backpropagation of error signals through the network, as required by the commonly used back error propagation algorithm, is avoided. It still guarantees, however, that the synaptic weights are updated in the error descent direction. With the apparent advantage of simpler optical implementation this learning rule is also shown by simulations to be computationally effective.
|Additional Information:||© Copyright 1992 Optical Society of America Received 30 September, 1991 This work was supported by the Defense Advanced Research Projects Agency and the U.S. Air Force Office of Scientific Research.|
|Subject Keywords:||Optical neural networks, anti-Hebbian local learning|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Archive Administrator|
|Deposited On:||25 May 2006|
|Last Modified:||26 Dec 2012 08:53|
Repository Staff Only: item control page