具有周期输入Hopfield型神经网络的全局渐近性质

On the Asymptotic Behavior of Hopfield Neural Network With Periodic Inputs

1.
Department of Physics, Hebei University of Technology, Tianjin 300130, P R China;
2.
Department of Mathematics, Shanghai University, Shang-hai 200436, P R China;
3.
Naval Submarine Academy, Qingdao 266071, P R China

摘要: 在不假定非线性激励函数有界和可微的条件下,应用Mawhin的重合度理论及Liapunov函数法给出一类具有周期输入的Hopfield型神经网络存在周期解及其全局指数稳定的充分条件。
- Hopfield网络 /
- 周期解 /
- 全局指数稳定 /
- 重合度 /
- Liapunov函数
Abstract: Without assuming the boundedness and differentiability of the nonlinear activation functions, the new sufficient conditions of the existence and the global exponential stability of periodic solutions for Hopfield neural network with periodic inputs are given by using Mawhin's coincidence degree theory and Liapunov's function method.
- Hopfield neural network /
- periodic solution /
- global exponential stability /
- coincidence degree /
- Liapunov function

[1]	Hopfield J J.Neural networks and physical system with emergent collective computational abilities[J].Proc Nat Academy Sci,1982,79(4):2554-2558.
[2]	Hopfield J J.Neurons with graded response have collective computational properties like those of two-state neurons[J].Proc Nat Academy Sci,1984,81(5):3088-3092.
[3]	廖晓昕.Hopfield型神经网络的稳定性[J].中国科学,1993,23(10):1025-1035.
[4]	梁学斌,吴立德.Hopfield型神经网络的全局指数稳定性[J].中国科学,1995,25(5):523-532.
[5]	Guan Z H,Chen G.On the equilibria,stability and instability of Hopfield neural networks[J].IEEE Trans Neural Networks,2000,11(2):534-539.
[6]	李铁成,王铎.一类带有周期输入的人工神经网络的渐近性质[J].高校应用数学学报,A辑,1997,12(1):25-28.
[7]	黄先开.具有时滞的Hopfield型神经网络的平稳周期振荡[J].应用数学和力学,1999,20(10):1040-1045.
[8]	Gao J.Periodic oscillation and exponential stability of delayed CNN[J].Physic Letter A,2000,270(3/4):157-163.
[9]	Gain R E,Mawhin J L.Coincidence Degree and Nonlinear Differential Equations[M].Lecture Note in Math,567,Berlin:Springer-Verlag,1977,40-41.
[10]	Yoshizawa T.Stability Theory by Liapunov's Second Method[M].Tokyo:The Math Soc of Japan,1996,165-169.

点击查看大图