Delay-derivative-dependent stability criterion for neural networks with probabilistic time-varying delay

In this article, based on Lyapunov–Krasovskii functional approach and improved delay-partitioning idea, a new sufficient condition is derived to guarantee a class of delayed neural networks to be asymptotically stable in the mean-square sense, in which the probabilistic time-varying delay is addressed. Together with general convex combination method, the criterion is presented via LMIs and its solvability heavily depends on the sizes of both time delay range and its derivative, which has wider application fields than those present ones. It can be shown by the numerical examples that our method reduces the conservatism much more effectively than earlier reported ones. Especially, the conservatism can be further decreased by thinning the delay intervals.