| Numéro |
J. Phys. France
Volume 51, Numéro 5, mars 1990
|
|
|---|---|---|
| Page(s) | 375 - 386 | |
| DOI | https://doi.org/10.1051/jphys:01990005105037500 | |
J. Phys. France 51, 375-386 (1990)
DOI: 10.1051/jphys:01990005105037500
1 Division of Chemistry, California Institute of Technology, Pasadena, CA 91125, U.S.A
2 Instituto de Fisica, Universidade Federal do Rio Grande do Sul, Caixa Postal 15051, 91500 Porto Alegre, RS, Brazil
0705M - Neural networks, fuzzy logic, artificial intelligence.
Key words
content addressable storage -- learning systems -- neural nets
DOI: 10.1051/jphys:01990005105037500
On the storage of correlated patterns in Hopfield's model
J.F. Fontanari1 et W.K. Theumann21 Division of Chemistry, California Institute of Technology, Pasadena, CA 91125, U.S.A
2 Instituto de Fisica, Universidade Federal do Rio Grande do Sul, Caixa Postal 15051, 91500 Porto Alegre, RS, Brazil
Abstract
The effects of storing p statistically independent but effectively correlated patterns in the Hopfield model of associative memory are studied. This leads us to propose a local learning rule which by enhancing the differences among the patterns allows the network to store them with an efficiency comparable to nonlocal learning rules.
Résumé
On étudie les effets de stockage de p patterns statistiquement indépendants mais effectivement corrélés dans le modèle de Hopfield de mémoire associative. Ceci conduit à proposer une règle d'apprentissage locale qui, en augmentant les différences entre patterns, conduit à une mémorisation d'efficacité comparable à celle des règles d'apprentissage non locales.
0705M - Neural networks, fuzzy logic, artificial intelligence.
Key words
content addressable storage -- learning systems -- neural nets