Almost Automorphic Solutions in Distribution Sense of Quaternion-Valued Stochastic Recurrent Neural Networks with Mixed Time-Varying Delays

Abstract

In this paper, we consider quaternion-valued stochastic recurrent neural networks with mixed time-varying delays by direct method. Base on the Banach fixed point theorem and stochastic analysis techniques, we obtain some sufficient conditions for the existence and global exponential stability of almost automorphic solutions in distribution sense for the neural networks. As special cases of our results, we also obtain the results about the existence and global exponential stability of almost automorphic solutions in distribution sense for complex-valued and real-valued stochastic neural networks. All of these results are new. Finally, we give numerical examples and simulations to illustrate the feasibility of our results.