Robust stability of switched uncertain stochastic recurrent neural networks with discrete and distributed delays

In this paper, some ideas of the switched systems are introduced into the field of neural networks and a class of switched uncertain stochastic recurrent neural networks (SUSRNNs) with discrete and distributed delays is investigated. In such neural networks, the features of switched systems, uncertain systems, stochastic systems, as well as time-delay systems are all taken into account. Based on the Lyapunov method and the stochastic analysis approach, some sufficient conditions are derived by means of linear matrix inequalities (LMIs) to guarantee the SUSRNNs to be globally robustly stable in the mean square. A simulation example is provided to illustrate the effectiveness of the proposed criteria.