ADAPTIVE WEIGHTED OUTER-PRODUCT LEARNING ASSOCIATIVE MEMORY

Associative-memory neural networks with adaptive weighted outer-produc t learning are proposed in this paper. For the correct recall of a fun damental memory (FM), a corresponding learning weight is attached and a parameter called signal-to-noise-ratio-gain (SNRG) is devised. The s ufficient conditions for the learning weights and the SNRG's are deriv ed. It is found both empirically and theoretically that the SNRG's hav e their own threshold values for correct recalls of the corresponding FM's. Based on the gradient-descent approach, several algorithms are c onstructed to adaptively find the optimal learning weights with refere nce to global- or local-error measure.