Stability of artificial neural networks with impulses

摘要: Sufficient conditions are obtained for the existence and asymptotic stability of a unique equilibrium Hopfield-type neural network with Lipschitzian activation functions without assuming their boundedness, monotonicity or differentiability subjected to impulsive state displacements at fixed instants time. Both continuous-time corresponding discrete-time networks considered. The sufficient do not restrict step-size appearing in discretization process these approach as tends zero those networks. terms parameters only easy verify; also when jumps absent results reduce non-impulsive systems.