Diluted neural networks with adapting and correlated synapses - art. no. 052904

We consider the dynamics of diluted neural networks with clipped and adapti ng synapses, Unlike previous studies, the learning rate is kept constant as the connectivity tends to infinity: the synapses evolve on a time scale in termediate between the quenched and annealing limits and all orders of syna ptic correlations must be taken into account. The dynamics is solved by mea n-field theory, the order parameter for synapses being a function. We descr ibe the effects, in the double dynamics, due to synaptic correlations.