1 October 1992 Novel two-layer neural multiassociative memory model
Author Affiliations +
Optical Engineering, 31(10), (1992). doi:10.1117/12.59883
Abstract
A novel, two-layer VLSI realizable, modular and expandable neural network multiassociative memory model is proposed. When compared to the Hopfield model, it requires 0(N) connections for N-bit long vectors; yields improved error correction and storage capabilities and a faster convergence rate, avoids the storage of false memories, and possesses analogies to biological neural networks.
Abdul Aziz Bhatti, Nalin Zhou, "Novel two-layer neural multiassociative memory model," Optical Engineering 31(10), (1 October 1992). http://dx.doi.org/10.1117/12.59883
JOURNAL ARTICLE
5 PAGES


SHARE
KEYWORDS
Neural networks

Computing systems

Content addressable memory

Very large scale integration

Performance modeling

Binary data

Computer networks

RELATED CONTENT


Back to Top