Spin glass model for a neural network : associative memories stored with unequal weights
Laura Viana · Journal de physique · 1988
A long range Spin Glass model is considered in order to study a neural network in which a finite number p of random patterns has been stored or memorised, each with unequal weight. This is done in order to reflect different degrees of training. The stability of these memories is studied and the possible existence of spurius stable states is investigated.