A Model of Chaotic Neural Network and Its Application
Gao Chun-tao · Harbin Ligong Daxue xuebao · 2010
We give a new kind of chaotic neuron model whose activation function is composed of the Sine function and Sigmoid function,through the bifurcation diagram and Lyapunov index calculation,an analysis of the characteristics of its dynamics.Based on this neuron model,we propose a new kind of transient chaotic neural network,and used to function combinational optimization problems.The simulation results show the efficiency of the proposed algorithm.