Global convergence of periodic solution of neural networks with discontinuous activation functions

被引:30
作者
Huang, Lihong [1 ]
Guo, Zhenyuan [1 ]
机构
[1] Hunan Univ, Coll Math & Econometr, Changsha 410082, Hunan, Peoples R China
基金
中国国家自然科学基金;
关键词
ASYMPTOTIC STABILITY; EXPONENTIAL STABILITY; DISTRIBUTED DELAYS; INCLUSIONS;
D O I
10.1016/j.chaos.2009.03.124
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
In this paper, without assuming boundedness and monotonicity of the activation functions, we establish some sufficient conditions ensuring the existence and global asymptotic stability of periodic solution of neural networks with discontinuous activation functions by using the Yoshizawa-like theorem and constructing proper Lyapunov function. The obtained results improve and extend previous works. (C) 2009 Elsevier Ltd. All rights reserved.
引用
收藏
页码:2351 / 2356
页数:6
相关论文
共 23 条