The time evolution of the distance between two random initial configurations subjected to the same thermal noise is used to study dynamical phase transitions in attractor neural networks trained by the Hebb rule. Numerical results are given for fully connected architectures, whereas, in the dilute case, both analytical and numerical outcomes are provided and a good agreement is shown to exist between the two sets of results.
DYNAMICAL PHASE-TRANSITIONS IN THE LITTLE-HOPFIELD MODEL
MARANGI C;PASQUARIELLO;
1994
Abstract
The time evolution of the distance between two random initial configurations subjected to the same thermal noise is used to study dynamical phase transitions in attractor neural networks trained by the Hebb rule. Numerical results are given for fully connected architectures, whereas, in the dilute case, both analytical and numerical outcomes are provided and a good agreement is shown to exist between the two sets of results.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.
