Robust Stability Of Interval Stochastic Neural Networks With Time-delays

In recent years, with the neural network widely applying in solving signal processing, optimizing solvers and pattern recognition, stability of stochastic neural networks received more and more attention and have obtained a series of results. In this thesis, robust stabil-ity of interval stochastic Hopfield neural networks and BAM neural networks are studied, which based on the theory of the Lyapunov stability and stochastic differential equation. The results are less conservative than some existing ones.In Chapter1, we introduce the theory of stability and some stochastic neural network models, then we present the development, research method and research progress of s-tochastic neural networks, finally the major work of this paper will be given.In Chapter2, we introduce some definitions of the stability of stochastic neural net-works.In Chapter3, new theory for robust stability of stochastic Hopfield neural networks has been derived using an approach combining the Lyapunov-Krasovskii functional with differ-ential, Ito equation and LMI techniques, and we present numerical examples to illustrate the validity of the new theory, by LMI toolbox.In Chapter4, this paper consider robust stability of stochastic BAM neural networks by constructing appropriate Lyapunov functionals, New criterions to satisfy robust stability of the system are gained, and a numerical example is given to show the validity of new criterions.In Chapter5, a summary of this paper is made, and the future research directions and forecasted.