混沌神经网络的Lyapunov指数与混沌区域

混沌神经网络有望用于诸如组合优化、信息搜索、模式识别等信息处理中.在Aihara等人依据生物学实验提出的混沌神经网络模型基础上,提出了混沌神经网络的最大Lyapunov指数的一种算法,通过最大Lyapunov指数的计算,研究神经网络的混沌区域以及参数对混沌区域的影响,这对混沌神经网络的混沌特性的研究及其混沌控制是十分有益的.

Lyapunov exponent and chaotic area distribution of a chaotic neural network

Chaotic neural networks is a potential technique of information processing such as combinatorial optimization, information search, pattern identification, etc. The chaotic neural network model based on biological experiments proposed by Aihara et al. exhibits very complex dynamics and proves its dynamics to be dependent on its parameters. An algorithm to calculate the largest Lyapunov exponent of the chaotic neural network is presented. Based on the largest Lyapunov exponent, we investigated the chaotic area distribution of the neural network and discuss its dependence on the parameters. The results are significant for investigating chaotic dynamics and controlling chaos in the chaotic neural network.