| 一类具M-P型非线性二元神经网络模型的周期解 |
| |
| 引用本文： | 刘开宇,侯振挺. 一类具M-P型非线性二元神经网络模型的周期解[J]. 国防科技大学学报, 2008, 30(4): 129-132 |
| |
| 作者姓名： | 刘开宇 侯振挺 |
| |
| 作者单位： | 湖南大学,数学与计量经济学院,湖南,长沙,410082;中南大学,数学科学与计算技术学院,湖南,长沙,410083;中南大学,数学科学与计算技术学院,湖南,长沙,410083 |
| |
| 摘 要： | 针对一类具M-P型非线性信号传输函数的二元神经网络模型,提出其初值振动时系统周期解的存在性问题.利用结式技巧,结合分析方法,建立了用结式表示的保证具振动初值系统周期解存在的充分条件,并通过例子说明零阈值情形周期解的稳定性.
|
| 关 键 词： | 神经网络 周期解 结式 |
| 收稿时间： | 2008-01-15 |
The Periodic Solution of a Neural Networks of Two Neurons with McCulloch-Pitts Nonlinearity |
| |
| LIU Kaiyu and HOU Zhenting. The Periodic Solution of a Neural Networks of Two Neurons with McCulloch-Pitts Nonlinearity[J]. Journal of National University of Defense Technology, 2008, 30(4): 129-132 |
| |
| Authors: | LIU Kaiyu and HOU Zhenting |
| |
| Affiliation: | LIU Kai-yu1,2,HOU Zhen-ting2 (1.College of Mathematics , Econometrics,Hunan Univ.,Changsha 410082,China,2.School of Mathematical Sciences , Computing Technology,Central South Univ.,Changsha 410083,China) |
| |
| Abstract: | The existence of the periodic solution for neural networks of two neurons with McCulloch-Pitts type signal functions and oscillatory initial value was proposed.By using the resultant techniques and method of analysis,resultant conditions which ensure the existence of periodic solution of the system with sign change of initial value were established.Furthermore,the stability of the periodic solution of the system with zero threshold was shown by example. |
| |
| Keywords: | neural network periodic solution resultant |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《国防科技大学学报》浏览原始摘要信息 |
|
点击此处可从《国防科技大学学报》下载免费的PDF全文 |
|
