Wandering Motion and Co-Operative Phenomena in a Chaotic Neural Network
Makoto Makishima, Toshihiro Shimizu · International Journal of Bifurcation and Chaos · 1998
A new model for associative memory with discrete time is proposed, in which chaos is taken into account and moreover the internal state of neurons depends on the past history. The network itself can search for minima of the energy successively by using the wandering motion and co-operative phenomena. By using this model we can clearly discuss the mechanism or the dynamics of the wandering motion and co-operative phenomena in a chaotic neural network to search for minima of the energy. It is shown that our model has great ability in the problem of associative memory.