Global Exponential Stability Of Three Kinds Of Neural Network Models

It is well known that it is an important theoretical issue for stability of neural network research. Modern life science research proves that:human consciousness, thought, awareness and cognition are closely related to people's brain neural systems. In order to simulate and reproduce the intelligence activities of human brain, people build a lot of artificial neural networks and their models. These models have important applications to the pattern recognition, signal processing, optimization and so on.This paper is concerned with global exponential stability of three types of neural network models. The major work is summarized below:In Chapter 1, we introduced background, significance of global exponential stability of neural networks and then gave arrangements of the main contents and results of the paper. Finally the symbols, definitions and lemmas which used in this article were illustrated.In Chapter 2, we studied the global exponential stability for a class of cellular neural networks with distributed delays. A sufficient condition ascertaining global exponential stability of the model's equilibrium point is obtained by constructing Lyapunov functions.In Chapter 3, we discussed the global exponential stability for a kind of BAM neural networks with delays. A sufficient condition ascertaining the existence, uniqueness and global exponential stability of the equilibrium point is given by using the fixed point theorem and constructing Lyapunov functions.In Chapter 4, we introduced the global exponential stability of BAM neural networks with delays and impulses. A sufficient condition ascertaining the existence, uniqueness and global exponential stability of the equilibrium point is obtained by using the fixed point theorem and constructing Lyapunov functions and differential inequality technique.