Learning rule for pattern recognition in Hopfield neuronal network

Authors
Andrecut, M
Citation
M. Andrecut, Learning rule for pattern recognition in Hopfield neuronal network, INT J MOD B, 12(32), 1998, pp. 3529-3537
Citations number
12
Categorie Soggetti
Apllied Physucs/Condensed Matter/Materiales Science
Journal title
INTERNATIONAL JOURNAL OF MODERN PHYSICS B
ISSN journal
02179792 → ACNP
Volume
12
Issue
32
Year of publication
1998
Pages
3529 - 3537
Database
ISI
SICI code
0217-9792(199812)12:32<3529:LRFPRI>2.0.ZU;2-W
Abstract
The Hopfield model of associative memories is reconsidered in terms of patt ern recognition in presence of noise. A learning rule for the investigated model is deduced. With the obtained learning rule the storing capacity of H opfield neuronal network is increased from alpha congruent to 0.14 to the o ptimal value alpha = 1. The pattern recognition capabilities of the model i ncrease exponentially with the dimensionality of the patterns.