A new model of chaotic neural network simulated annealin and its application

Unlike the gradient descent neural network, the chaotic neural networks have more complex dynamics. A kind of chaotic neural network is presented, which combines stochastic with deterministic property to introduce chaos mechanism into Hopfield neural network to coarsely search the optimum under chaotic dynamics and control the chaotic dynamics by annealing strategy to perform inverse bifurcation and disappear. After that, the gradient property of HNN is used to reach stable point. Simulation results show that such an algorithm can avoid getting stuck in local minima.