The Hopfield-type Memory Without Catastrophic Forgetting

Earlier some modifications of the Hebb matrix were proposed to eliminate the memory destruction [3]-[7]. As the result of such modifications an unlimited number of random patterns can be fearlessly written down into matrix elements one by one. However, the memory of the network is restricted. If as previously the maximum number of recognized patterns is denoted by Abstract—We analyzed a Hopfield-like model of artificial memory that reproduces some features of the human memory. They are: a) the ability to absorb new information when working; b) the memorized patterns are only a small part of a set of patterns that are written down in connection matrix; c) the more the pattern was shown during the learning process, the better the quality of its recognition. We used the Hebb rule, but each pattern was supplied with its own weight. The weigh