A Back Propoagation Model with Periodic Chaos Neurons

摘要

In this paper we report a back propagation scheme with a periodic chaos neuron model. In contrast to the conventional model, each neuron in the network may involve the chaotic behaviour in the learning process. In practice one may confirm that there exists a close relation between the learning process and the chaotic dynamics, and that chaotic dynamics promotes the learning speed with a certain succerss rate as seen in the conventional nonchaotic models. The parameter related to the chaos will be controlled to drive the system from a chaos to a nonchaos through the learning process to assure the high success rate.