Absolute exponential stability analysis of recurrent neural networks with generalized activation function: An LMI approach

This paper is devoted to study the absolute exponential stability of recurrent neural network with novel generalized activation function, which is recently proposed in my previous paper. By integrating Lyapunov stability theory and LMI approach, the stability criterion is derived, which is in form of LMI with slack variables. It may enlarge the range in selecting neural networks¿ parameters. Moreover, the stability criteria become less conservative than my previous paper.