Stability analysis of delayed neural networks: An auxiliary matrix-based technique

This paper revisits the problem of stability analysis for delayed neural networks (DNNs). By introducing a set of auxiliary vectors and slack matrices, an auxiliary matrix-based integral inequality (AMBII) is presented. The auxiliary matrix is composed of auxiliary vectors, slack matrix and time-varying delay. It can make a trade off between conservatism and complexity. By using AMBII, a less conservative stability criterion is obtained for DNNs in terms of linear matrix inequalities (LMIs). The effectiveness of the stability condition can be demonstrated by illustrating a numerical example.