Chaotic Neural Network with Initial Value Reassigned and Its Application
In this paper, three existing chaotic neural network models are reviewed and the searching ability of these models is analyzed, a novel chaotic neural network with varying initial value is proposed to solve problems of the lower convergence rate and long searching time in the existing method. It is different from the other modified chaotic neural networks in the aspect that it seeks the better initial value that can lead to the global optimized solution in limited steps by means of chaotic iteration instead of enlarging the annealing time or modifying annealing parameters. The new method can get the increasing convergence rate and the decreasing searching time. The controlled numerical experiments with the Travel Salesman Problems (TSP) show that the proposed method has better global searching ability.
KeywordsConvergence Rate Travel Salesman Problem Global Optimal Solution Chaotic State Hopfield Neural Network
Unable to display preview. Download preview PDF.
- 2.Aihara, K.: Bifurcation Phenomena in Nonlinear Systems and Theory of Dynamical Systems. In: Kawakami, H. (ed.) Chaotic Neural Network, pp. 143–161. World Scientific, Singapore (1990)Google Scholar
- 4.Chen, L., Aihara, K.: Chaotic Simulated Annealing by a Neural Network Model with Transient Chaos. IEEE Transactions on Neural Networks 8(6), 915–930 (1995)Google Scholar
- 8.Xiu, C., Liu, X.: A Novel Chaotic Neural Network and Its Application. Acta Electronica Sinice 33(5), 868–870 (2005)Google Scholar