CAPACITIES OF MULTICONNECTED MEMORY MODELS.

D. Horn*, M. Usher

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

We study generalizations of the Hopfield model for associative memory which contain interactions of R spins with one another and allow for different weights for input patterns. Using probabilistic considerations we show that stability criteria lead to capacities which increase like powers of N**R** minus **1. Investigating numerically the basins of attraction we find behavior which agrees with theoretical expectations. We introduce the more stringent definition of 'converage-capacity' by requiring the whole phase-space to be covered by the basins of attraction of the input patterns. Even under these conditions we find large numbers of patterns which can be used to design an output spectrum by varying the input weights.

Original languageEnglish
Pages (from-to)389-395
Number of pages7
JournalJournal De Physique
Volume49
Issue number3
DOIs
StatePublished - 1988

Fingerprint

Dive into the research topics of 'CAPACITIES OF MULTICONNECTED MEMORY MODELS.'. Together they form a unique fingerprint.

Cite this