A study of exponential stability for stochastic delayed neural networks

This paper is concerned with analyzing mean square exponential stability of stochastic delayed neural networks subject to parametric uncertainties. The discretized Lyapunov functional technique is first utilized to construct a new Lyapunov functional in order to effectively deal with the time-varying delay. Then the free-weighting matrix technique and the convex combination method are used to establish a new delay-dependent mean square exponential stability criterion for uncertain stochastic delayed neural networks. The usefulness of the new theoretical findings is further demonstrated by numerical results.