摘要
将动力学和统计学方法结合,分析了一类具有时滞的加权小世界环状神经网络的稳定性情况.从数学期望的角度找到临界值使网络关于平衡状态渐进稳定.对比神经元个数的不同取值对分析结果的影响,从而形成误差分析.得出含有时滞的加权小世界神经网络要比完全连接的规则网络更容易稳定这一结论.
This paper presents an analysis of the stability of a weighted small-world ring neural network with time delay by combining dynamical and statistical methods.We can find critical values to make the network asymptotically stable about the origin in the sense of mathematical expectation.By comparison of different values for n which is the number of neurons with the analysis of the influence,thus forming error analysis.We conclude a weighted small-world network with time delay is more stable than the fully connected rule network.
作者
盛夏
徐旭
SHENG Xia;XU Xu(Department of Fundamental Science,The City College Of Jilin Jianzhu University,Changchun 130014;College of Mathematics,Jilin University,Changchun 130012,China)
出处
《数学的实践与认识》
北大核心
2019年第6期217-222,共6页
Mathematics in Practice and Theory
基金
国家自然科学基金(11372117)
关键词
小世界网络
时滞
动力学
统计学
稳定性
small-word network
time delay
dynamics
statistics
stability
