1 May 1992 Derivation of neural network models and their computational circuits for associative memory
Author Affiliations +
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, Eung Gi Paek, Paul F. Liao, Paul F. Liao, Hamid Gharavi, Hamid Gharavi, } "Derivation of neural network models and their computational circuits for associative memory," Optical Engineering 31(5), (1 May 1992). https://doi.org/10.1117/12.56162 . Submission:
JOURNAL ARTICLE
10 PAGES


SHARE
RELATED CONTENT

Multiresolution neural networks
Proceedings of SPIE (March 14 1994)
Feedback neural network for pattern recognition
Proceedings of SPIE (March 08 1999)
Novel model of linear associative memory
Proceedings of SPIE (June 30 1992)
Image preprocessing by neuronlike algorithms
Proceedings of SPIE (March 31 1998)
Genetic algorithms applied to optics and engineering
Proceedings of SPIE (February 09 2006)

Back to Top