Stability Of Uncertain Neural Network

This paper mainly researches the stability of neural network with uncertain distur-bance. Neural networks are susceptible to external influence such as stochastic factors and uncertain factors, because of the complexity of itself. A number of researches have been reported on stability of Hopfield neural networks at home and abroad. Besides, there are also some achievements on the stability of neural networks with random disturbance. However, in practice, apart from stochastic factors, we need to concern uncertain factors to analyze the stability of neural networks. In this paper, motivated by the study on the stability of stochastic neural networks, we design methods for uncertain neural network stability studies based on uncertainty theory established by professor Baoding Liu. Stability of uncertain neural network is actually stability of uncertain different equations. But, in earlier studies on stability of uncertain differ-ential equation, most researchers introduced stability method only by definitions of stability. In this paper, we will establish Lyapunov method to study stability of un-certain Hopfield neural network. Then we will give two conclusions for stability based on different structure of neural network. Finally we will obtain some stability results when the uncertain disturbance is a special case of linear. In addition, we will discuss the stability of two neural networks with special structure and special disturbance.