Higher Order Neurodynamics of Associative Memory for Sequential Patterns

摘要

This paper describes higher order neurodynamics of associative memory for sequential patterns using a statistical method. First, the statistical analysis of direct correlations between the cross talk noise terms for higher order neural networks is made. Further, it is shown that storage capacities for k = 1, 2 and 3 dimensional cases are 0.263n, 0.207(n 2) and 0.180(n 3), respectively, where n is the number of neurons and (n k) means the combination of k from n. The result for the one dimensional case is in fairly general agreement with Meir's result, 0.269n, obtained by the replica theory.