The authors consider neural networks trained by the symmetric Edinburgh algorithm at finite temperature. They show that, while the introduction of the thermal noise affects the dynamics in a computable way, it does not change the domains of attraction of stored patterns.
DOMAINS OF ATTRACTION OF NEURAL NETWORKS AT FINITE TEMPERATURE
PASQUARIELLO G
1991
Abstract
The authors consider neural networks trained by the symmetric Edinburgh algorithm at finite temperature. They show that, while the introduction of the thermal noise affects the dynamics in a computable way, it does not change the domains of attraction of stored patterns.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
