ROBUSTNESS AND PERTURBATION ANALYSIS OF A CLASS OF NONLINEAR-SYSTEMS WITH APPLICATIONS TO NEURAL NETWORKS

In this paper we study the robustness properties of a large class of n onlinear systems by addressing the following question: given a nonline ar system with specified asymptotically stable equilibria, under what conditions will a perturbed model of the system possess asymptotically stable equilibria that are close (in distance) to the asymptotically stable equilibria of the unperturbed system? In arriving at our result s, we establish robustness stability results for the perturbed systems considered, and we determine conditions that ensure the existence of asymptotically stable equilibria of the perturbed system that are near the asymptotically stable equilibria of the original unperturbed syst em. These results involve quantitative estimates of the distance betwe en the corresponding equilibrium points of the unperturbed and perturb ed systems. We apply the above results in the qualitative analysis of a large class of artificial neural networks.