1 May 1992 Derivation of neural network models and their computational circuits for associative memory
Author Affiliations +
Optical Engineering, 31(5), (1992). doi:10.1117/12.56162
Abstract
Neural network models for associative memory are derived independently on the basis of an optimization principle without resort to any assumptions related to biological principles. All the features of the Hopfield model, such as the updating rule with nonlinear threshold, the outer product algorithm, the symmetric and zero-diagonal interconnection matrix, and asynchronous timing, are automatically derived from a simple optimization principle for bipolar and binary variables. The derivation is extended to generate higher order models that have higher storage capacity and better convergence. The computational circuits to implement the neural network models are also derived naturally from the same principle. Various optical implementations of the computational circuits are also described.
Eung Gi Paek, Paul F. Liao, Hamid Gharavi, "Derivation of neural network models and their computational circuits for associative memory," Optical Engineering 31(5), (1 May 1992). http://dx.doi.org/10.1117/12.56162
JOURNAL ARTICLE
10 PAGES


SHARE
KEYWORDS
Content addressable memory

Neural networks

Optimization (mathematics)

Binary data

Algorithm development

Computing systems

Chemical elements

RELATED CONTENT

Development Of Autonomous Systems
Proceedings of SPIE (March 21 1989)
History of a stochastic growth model
Proceedings of SPIE (February 01 1998)
Detecting change in images with parallax
Proceedings of SPIE (May 07 2007)
Mobile code security
Proceedings of SPIE (November 08 2001)

Back to Top