一类具M-P型非线性神经网络模型的渐近性与指数稳定  

Asymptotic Behavior and Exponential Stability of Solutions of Neural Network with McCulloch-Pitts Nonlinearity

作　　者：刘开宇[1] 张弘强[2] 

机构地区：[1]中南大学数学科学与计算技术学院 [2]长沙理工大学数学与计算科学学院,中国长沙410076

出　　处：《湖南师范大学自然科学学报》2007年第2期6-10,共5页Journal of Natural Science of Hunan Normal University

基　　金：国家自然科学基金资助项目(10571032)

摘　　要：研究了二元非线性时滞神经网络模型.获得了大阈值及临界阈值情形的渐近行为与全局指数稳定.特别是对临界情形建立了解趋于各平衡点的充要条件,其结果对相关文献的结论作了较大推广.Asymptotic behavior and exponential stability of solutions of neural networks with McCulloch-Pitts nonlinearity are investigated. We discuss asymptotic behaviors and global exponential stability in the different thresholds cases. In particular, some new necessary and sufficient conditions are extablished in the critical case. This resuhs generalize the previous results given in the literature.

关 键 词：神经网络 渐近性 全局指数稳定 

分 类 号：O175.12[理学—数学]