Title :
Global asymptotic stability of discrete-time recurrent neural networks
Author :
Hu, Sanqing ; Wang, Jun
Author_Institution :
Dept. of Autom. & Comput.-Aided Eng., Chinese Univ. of Hong Kong, Shatin, China
Abstract :
This paper presents new analytical results on the global asymptotic stability for the equilibrium states of a general class of discrete-time recurrent neural networks (DTRNNs) described by using a set of nonlinear difference equations. We provide a few sufficient conditions for the global asymptotic stability of DTRNNs. The resulting criteria include diagonal stability and nondiagonal stability. These stability conditions are less restrictive than the existing ones in the literature
Keywords :
asymptotic stability; difference equations; recurrent neural nets; asymptotic stability; diagonal stability; discrete-time neural networks; equilibrium states; nondiagonal stability; nonlinear difference equations; recurrent neural networks; sufficient conditions; Asymptotic stability; Automation; Digital filters; Lyapunov method; Neural networks; Nonlinear equations; Recurrent neural networks; Stability criteria; State feedback; Sufficient conditions;
Conference_Titel :
Decision and Control, 2000. Proceedings of the 39th IEEE Conference on
Conference_Location :
Sydney, NSW
Print_ISBN :
0-7803-6638-7
DOI :
10.1109/CDC.2000.912881
