Ruoxia Li, Jinde Cao, Ahmad Alsaedi, Fuad Alsaadi Stability analysis of fractional-order delayed neural networks

Abstract

At the beginning, a class of fractional-order delayed neural networks were employed. It is known that the active functions in a target model may be Lipschitz continuous, while some others may also possessing inverse Lipschitz properties. Based upon the topological degree theory, nonsmooth analysis, as well as nonlinear measure method, several novel sufficient conditions are established towards the existence as well as uniqueness of the equilibrium point, which are voiced in terms of linear matrix inequalities (LMIs). Furthermore, the stability analysis is also attached. One numerical example and its simulations are presented to illustrate the theoretical findings.

Keywords

• fractional-order neural network
• inverse Lipschitz neuron activations
• topological degree theory
• stability analysis