AN AUTOCORRELATION ASSOCIATIVE NEURAL-NETWORK WITH SELF-FEEDBACKS

In this article, the autocorrelation associative neural network that i s one of well-known applications of neural networks is improved to ext end its capacity and error correcting ability. Our approach of the imp rovement is based on the consideration that negative self-feedbacks re move spurious states. Therefore, we propose a method to determine the self-feedbacks as small as possible within the range that all stored p atterns are stable. A state transition rule that enables to escape osc illation is also presented because the method has a possibility of fal ling into oscillation. The efficiency of the method is confirmed by me ans of some computer simulations.