A new stability criterion of stochastic neural networks with delays

This paper investigates the problem of mean-square asymptotic stability of uncertain neural networks with time-varying delay and stochastic noise. Based on generalized Finsler lemma and the linear matrix inequality (LMI) optimization technique, an improved delay-dependent stability criterion is developed. It is shown that the new stability criterion is less conservative and less computationally complex than the existing stability conditions. A numerical example is presented to substantiate the effectiveness of the theoretical results.