BP Neural Network Based on Chaos Simulated Annealing

Combining chaos simulated annealing with gradient searching, a novel learning method for BP neural network is proposed. In the process of learning, the neural network will experience coarsely search and elaborately search. Taking advantage of the randomicity property and ergodicity property of chaos, the neural network gets the ability of escaping from the local minimized point and getting the global optimal point. Numbers of simulation experiments are realized by using the method in this paper. The simulation results prove the validity of the algorithm.