Stability analysis for the generalized Hopfield neural networks with multi-level activation functions

被引：4
作者：
Liu, Yiguang [1 ,2 ]
You, Zhisheng [1 ,2 ]
机构：
[1] Sichuan Univ, Inst Image & Graph, Key Lab Fundamental Synthet Vis Graph & Image Nat, Chengdu 610064, Peoples R China
[2] Sichuan Univ, Sch Comp Sci & Engn, Chengdu 610064, Peoples R China
来源：
NEUROCOMPUTING | 2008年 / 71卷 / 16-18期
基金：
中国国家自然科学基金;
关键词：
Hopfield neural networks; Multi-level activation function; Local or global stability; Complete stability;
D O I：
10.1016/j.neucom.2008.06.002
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
Using the continuity property of neuron state variables and Lyapunov functional, this paper religiously gives sufficient conditions ensuring the equilibrium number, local stable state number, global stability and complete stability of the Hopfield neural networks with multi-level activation functions. Compared with the known relevant literatures, the obtained results are more precise. Finally, three examples are employed to illustrate the obtained criteria. (C) 2008 Elsevier B.V. All rights reserved.
引用
收藏
页码：3595 / 3601
页数：7
相关论文
共 15 条
[1]   A note on the global stability of dynamical neural networks [J].
Arik, S .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS, 2002, 49 (04) :502-504
[2]   Hierarchical content classification and script determination for automatic document image processing [J].
Chi, Z ;
Wang, Q ;
Siu, WC .
PATTERN RECOGNITION, 2003, 36 (11) :2483-2500
[3]   On equilibria, stability, and instability of hopfield neural networks [J].
Guan, ZH ;
Chen, GR ;
Qin, Y .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2000, 11 (02) :534-540
[4]  
HU S, 1994, IEEE T NEURAL NETWOR, V5, P719
[5]   Stability analysis of Hopfield-type neural networks [J].
Juang, JC .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1999, 10 (06) :1366-1374
[6]   An adaptive resource-allocating network for automated detection, segmentation, and classification of breast cancer nuclei topic area: Image processing and recognition [J].
Lee, KM ;
Street, WN .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2003, 14 (03) :680-687
[7]   A new result on the global convergence of Hopfield neural networks [J].
Liu, XW ;
Chen, TP .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS, 2002, 49 (10) :1514-1516
[8]   On stability of disturbed Hopfield neural networks with time delays [J].
Liu, YG ;
You, ZS ;
Cao, LP .
NEUROCOMPUTING, 2006, 69 (7-9) :941-948
[9]   Dynamical behaviors of Hopfield neural network with multilevel activation functions [J].
Liu, YG ;
You, ZS ;
Cao, LP .
CHAOS SOLITONS & FRACTALS, 2005, 25 (05) :1141-1153
[10]   Multi-stability and almost periodic solutions of a class of recurrent neural networks [J].
Liu, Yiguang ;
You, Zhisheng .
CHAOS SOLITONS & FRACTALS, 2007, 33 (02) :554-563
12