Output convergence analysis of continuous-time recurrent neural networks

Abstract

This paper discusses the global output convergence of a class of continuous-time recurrent neural networks with globally Lipschitz continuous and monotone nondecreasing activation functions and locally Lipschitz continuous time-varying thresholds. We establish several sufficient conditions to guarantee the global output convergence for this class of neural networks. The present results do not require symmetry in the connection weight matrix. These convergence results are useful in the design of the recurrent neural networks with time-varying thresholds.