| 不连续激活函数的时滞神经网络的多稳定性 |
| |
| 引用本文： | 陈晓丰,宋乾坤.不连续激活函数的时滞神经网络的多稳定性[J].应用数学和力学,2013,34(4):383-398. |
| |
| 作者姓名： | 陈晓丰 宋乾坤 |
| |
| 作者单位： | 重庆交通大学 理学院,重庆 400074 |
| |
| 基金项目： | 国家自然科学基金资助项目 |
| |
| 摘 要： | 研究了不连续激活函数的时滞神经网络的多稳定性问题，在所研究的神经网络中，激活函数并不需要是连续的和单调的．给出了判断该神经网络多个平衡点存在及局部指数稳定的充分条件．最后，给出了两个数值仿真例子来验证本文获得结果的有效性和较小的保守性．
|
| 关 键 词： | 神经网络 不连续激活函数 多重稳定 指数稳定 |
| 收稿时间： | 2013-03-18 |
Multistability of Delayed Neural Networks With Discontinuous Activations |
| |
| Affiliation: | Department of Mathematics, Chongqing Jiaotong University,Chongqing 400074, P.R.China |
| |
| Abstract: | The problem on the multistability was investigated for delayed neural networks with discontinuous activations. For the neural networks under studied,the traditional assumptions on the continuity and the monotonicity of the activation functions were not required. Several sufficient conditions for checking the coexistence and local exponential stability of equilibria for the considered neural networks were given. Finally, two numerical examples were given to show the effectiveness and less conservatism of the proposed criteria. |
| |
| Keywords: | |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《应用数学和力学》浏览原始摘要信息 |
|
点击此处可从《应用数学和力学》下载全文 |
|
