Mean-square exponential input-to-state stability of stochastic recurrent neural networks with multi-proportional delays

This paper investigates mean-square exponential input-to-state stability of stochastic recurrent neural networks with multi-proportional delays. Here, we study the proportional delay, which is a kind of unbounded time-varying delay in stochastic recurrent neural networks, by employing Lyapunov-Krasovskii functional, stochastic analysis theory and It o^′s formula. A new stability criterion about the mean-square exponential input-to-state stability, which is different from the traditional stability criteria, is presented. In addition, the new proposed criterion easy to verify and less conservation than earlier publications about mean-square exponential input-to-state stability of stochastic recurrent neural networks. Finally, several examples and their simulations are given to illustrate the correctness and effectiveness of the theoretical results.