Exponential stability of Cohen-Grossberg neural networks with a general class of activation functions
文献类型:期刊论文
作者 | Wan, AH; Wang, MS; Peng, J![]() ![]() |
刊名 | PHYSICS LETTERS A
![]() |
出版日期 | 2006-01-30 |
卷号 | 350期号:1-2页码:96-102 |
关键词 | Cohen-Grossberg neural networks exponential stability exponential decay estimate nonlinear measure |
英文摘要 | In this Letter, the dynamics of Cohen-Grossberg neural networks model are investigated. The activation functions are only assumed to be Lipschitz continuous, which provide a much wider application domain for neural networks than the previous results. By means of the extended nonlinear measure approach, new and relaxed sufficient conditions for the existence, uniqueness and global exponential stability of equilibrium of the neural networks are obtained. Moreover, an estimate for the exponential convergence rate of the neural networks is precisely characterized. Our results improve those existing ones. (c) 2005 Elsevier B.V. All rights reserved. |
WOS标题词 | Science & Technology ; Physical Sciences |
类目[WOS] | Physics, Multidisciplinary |
研究领域[WOS] | Physics |
关键词[WOS] | ABSOLUTE STABILITY |
收录类别 | SCI |
语种 | 英语 |
WOS记录号 | WOS:000234935400017 |
公开日期 | 2015-12-24 |
源URL | [http://ir.ia.ac.cn/handle/173211/9371] ![]() |
专题 | 自动化研究所_09年以前成果 |
作者单位 | 1.Chinese Acad Sci, Inst Automat, Beijing 100080, Peoples R China 2.Xian Jiaotong Univ, Fac Sci, Inst Informat & Syst Sci, Xian 710049, Peoples R China |
推荐引用方式 GB/T 7714 | Wan, AH,Wang, MS,Peng, J,et al. Exponential stability of Cohen-Grossberg neural networks with a general class of activation functions[J]. PHYSICS LETTERS A,2006,350(1-2):96-102. |
APA | Wan, AH,Wang, MS,Peng, J,&Qiao, H.(2006).Exponential stability of Cohen-Grossberg neural networks with a general class of activation functions.PHYSICS LETTERS A,350(1-2),96-102. |
MLA | Wan, AH,et al."Exponential stability of Cohen-Grossberg neural networks with a general class of activation functions".PHYSICS LETTERS A 350.1-2(2006):96-102. |
入库方式: OAI收割
来源:自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。