Effects of the chaotic noise on the performance of a neural network model for optimization problems

• DOI: 10.1103/physreve.51.r2693
• Publication Date: 2002-07-27T01:07:11Z
• Authors: Yoshinori Hayakawa; Atsushi Marumoto; Yasuji Sawada
• Keywords: 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

We studied effects of chaos added to the dynamics of a neural network model. By numerical simulations, we found the neural network with forcing by chaotic noise operated very efficiently to solve an optimization problem. We also showed that short time correlation of chaos was relevant to the dynamics of the network and it could work effectively for global minima search.