We investigate the retrieval properties of fully connected saturated neural networks trained by the Edinburgh algorithm with correlated patterns {?i}. We evaluate the effect of the correlation on the overlap between {?i} and the network configuration after one time step and on the basins of attraction. We also show that the introduction of the thermal noise does not change the domains of attraction of the correlated memories.
Dynamics of fully connected neural network with sign constraints
C Marangi;G Pasquariello
1992
Abstract
We investigate the retrieval properties of fully connected saturated neural networks trained by the Edinburgh algorithm with correlated patterns {?i}. We evaluate the effect of the correlation on the overlap between {?i} and the network configuration after one time step and on the basins of attraction. We also show that the introduction of the thermal noise does not change the domains of attraction of the correlated memories.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.
