Dynamical stability analysis of multiple equilibrium points in time-varying delayed recurrent neural networks with discontinuous activation functions

被引：67
作者：
Huang, Yujiao [1 ]
Zhang, Huaguang [1 ]
Wang, Zhanshan [1 ]
机构：
[1] Northeastern Univ, Sch Informat Sci & Engn, Shenyang 110819, Liaoning, Peoples R China
来源：
NEUROCOMPUTING | 2012年 / 91卷
基金：
中国国家自然科学基金;
关键词：
Recurrent neural network; Discontinuous activation function; Multistability; ROBUST EXPONENTIAL STABILITY; GLOBAL CONVERGENCE; ASSOCIATIVE MEMORY; MULTISTABILITY; MULTIPERIODICITY; ATTRACTIVITY;
D O I：
10.1016/j.neucom.2012.02.016
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
This paper is concerned with the dynamical stability analysis of multiple equilibrium points in recurrent neural networks with time-varying delays and discontinuous activation functions. Based on the decomposition of state space, some sufficient conditions for the existence of multiple equilibrium points are established, which ensure that n-dimensional recurrent neural networks with k-level discontinuous activation functions can have k(n) equilibrium points. Under these conditions, the equilibrium points are locally exponentially stable. Moreover, some conditions for the existence of sets of stable equilibrium points and unstable equilibrium points are derived for recurrent neural networks without delay and with discontinuous activation functions. Finally, three examples are given to illustrate the effectiveness of the results. (C) 2012 Elsevier B.V. All rights reserved.
引用
收藏
页码：21 / 28
页数：8
相关论文
共 39 条
[1]   Multistability and multiperiodicity of delayed Cohen-Grossberg neural networks with a general class of activation functions [J].
Cao, Jinde ;
Feng, Gang ;
Wang, Yanyan .
PHYSICA D-NONLINEAR PHENOMENA, 2008, 237 (13) :1734-1749
[2]   New theorems on global convergence of some dynamical systems [J].
Chen, TP ;
Amari, S .
NEURAL NETWORKS, 2001, 14 (03) :251-255
[3]   Multistability and convergence in delayed neural networks [J].
Cheng, Chang-Yuan ;
Lin, Kuang-Hui ;
Shih, Chih-Wen .
PHYSICA D-NONLINEAR PHENOMENA, 2007, 225 (01) :61-74
[4]   Multistability in recurrent neural networks [J].
Cheng, Chang-Yuan ;
Lin, Kuang-Hui ;
Shih, Chih-Wen .
SIAM JOURNAL ON APPLIED MATHEMATICS, 2006, 66 (04) :1301-1320
[5]  
Filippov A.F., 1988, Differential Equations with Discontinuous Righthand Sides, P3
[6]   Generalized Lyapunov approach for convergence of neural networks with discontinuous or non-Lipschitz activations [J].
Forti, A ;
Grazzini, A ;
Nistri, P ;
Pancioni, L .
PHYSICA D-NONLINEAR PHENOMENA, 2006, 214 (01) :88-99
[7]   Global exponential stability and global convergence in finite time of delayed neural networks with infinite gain [J].
Forti, M ;
Nistri, P ;
Papini, D .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2005, 16 (06) :1449-1463
[8]   Global convergence of neural networks with discontinuous neuron activations [J].
Forti, M ;
Nistri, P .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2003, 50 (11) :1421-1435
[9]   M-matrices and global convergence of discontinuous neural networks [J].
Forti, Mauro .
INTERNATIONAL JOURNAL OF CIRCUIT THEORY AND APPLICATIONS, 2007, 35 (02) :105-130
[10]   Multistability of neural networks with discontinuous activation function [J].
Huang, Gan ;
Cao, Jinde .
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION, 2008, 13 (10) :2279-2289
1234