连续型BAM神经网络的指数稳定性

EXPONENTIAL STABILITY OF CONTINUOUS BAM NEURAL NETWORK

首先将连续型双向联想记忆神经网络转化成一个特殊的 Ｈｏｐｆｉｅｌｄ网络模型．在此基础上，对连续ＢＡＭ神经网络的指数稳定性进行了新的分析，证明了神经网络连接权矩阵在给定的约束条件下有唯一平衡点．所做的分析可以用于设计全局指数稳定的神经网络．

In this paper, the continuous bidirectional associative memory(BAM) neural networks can be considered as a special Hopfield network model. A novel exponential stability analysis is presented for the equilibrium points of continuous BAM neural networks. A constraint condition on the connection matrix has been found under which the neural network has a unique equilibrium point. The analysis in this paper can be used to design globally exponentially stable neural networks.