Improved dynamical model for neural network and the stabilization of associative memory

A new neural network model with more biological background is presented. Besides the usual properties of internal potential, accumulation and output, the new model is endowed with some dynamical factors such as local potential accumulation, adaptive variation of the threshold, the lateral links among the corresponding neurons as well as the adaptive change of the link weights in the evolution of the network. The typical dynamics and characteristics of associative memory can also be demonstrated through varying the values of the parameters in the neural network.