OPTIMUM LEARNING FOR BIDIRECTIONAL ASSOCIATIVE MEMORY IN THE SENSE OFCAPACITY

Borrowing the idea of Perceptron, Bidirectional Learning (BL) is propo sed here to enhance the recall performance of Bidirectional Associativ e Memory (BAM). By modifying the proof of convergence of Perceptron, w e have proved that BL yields one of the solution connection matrices w ithin a finite number of iterations (if the solutions exist). Accordin g to the above convergence of BL, the capacity of BAM with BL is large r than or equal to that with any other learning rule. Hence, BL can be considered as an optimum learning rule for BAM in the sense of capaci ty. Simulations show that BL greatly improves the capacity and the err or correction capability of BAM.