Exponential convergence estimates for neural networks with multiple delays
Tianguang Chu, Zhaolin Wang, Long Wang · IEEE Transactions on Circuits and Systems I Fundamental Theory and Applications · 2002
Componentwise estimates of (global) exponential convergence are obtained for a class of neural networks with multiple delays by com- parison method for delay differential equations. The method is simple and straightforward and does not resort to any Lyapunov functional. The main result shows explicitly the effect of time delays on exponential decay rate of the networks and is hence of practical interest in designing a fast and stable neural network. Compared with some existing results for stability of the Hopfield network model and delayed cellular neural network model via Lyapunov method, the present results can provide further detailed charac- terization of the convergence behavior for the systems under the same or similar conditions.