Behavior of interactive neural networks as associative memories

We analyze the main characteristics of neural networks interacting between them and making a system. The subnets are of Hopfield type. The synoptic intensities of the connections belonging to each of the sub-nets and those that connect both are established by Hebb´s rule. We study the application and influence of an unlearning procedure that modifies the intensity of the synoptic connection in the diverse regions of the system. The retrieval of patterns with different degree of correlation is analyzed. The retrieval performance is studied for this particular interacting system, determining the influence of the connectivity inter and intra sub-net and the correlation of the patterns recovered