Almost Periodic Solution And Global Exponential Stability Of Two Kinds Of Neural Networks

In this paper,the global exponential stability and almost periodic solutions of the Clifford-value leaky integrator echo state neural network with time delay and the glob-al exponential stability and almost periodic solutions of the Quaternion-valued neural network with proportional time delay are studied.Firstly,by means of the fixed point principle of Banach spatial compression mapping,the sufficient conditions for the exis-tence of almost periodic solutions of the echo state neural network are obtained,and the global exponential stability of the solutions is proved by the application of a proof by con-tradiction.Secondly,according to the fixed point principle of Banach spatial compression mapping,a sufficient condition for the existence of almost periodic solutions of a quater-nion numerical neural network with proport,ional time delay is obtained,and the global exponential stability of the solution is studied by using Halanay inequality.Thirdly,we illustrate the validity of the research results by examples.Finally,the conclusion of this paper is given.