An Overview of the Stability Analysis of Recurrent Neural Networks With Multiple Equilibria

被引：44
作者：
Liu, Peng [1 ,2 ]
Wang, Jun [3 ]
Zeng, Zhigang [4 ,5 ]
机构：
[1] Zhengzhou Univ Light Ind, Sch Elect & Informat Engn, Zhengzhou 450002, Peoples R China
[2] Henan Key Lab Informat Based Elect Appliances, Zhengzhou 450002, Peoples R China
[3] City Univ Hong Kong, Sch Data Sci, Dept Comp Sci, Hong Kong, Peoples R China
[4] Huazhong Univ Sci & Technol, Sch Artificial Intelligence & Automat, Wuhan 430074, Peoples R China
[5] Educ Minist China, Key Lab Image Proc & Intelligent Control, Wuhan 430074, Peoples R China
来源：
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS | 2023年 / 34卷 / 03期
基金：
中国国家自然科学基金;
关键词：
Stability criteria; Switches; Recurrent neural networks; Mathematical model; Stairs; Delay effects; Analytical models; Complete stability; multiple equilibria; multistability; recurrent neural networks (RNNs); LINEAR ACTIVATION FUNCTIONS; LIMIT SET DICHOTOMY; MU-STABILITY; MULTISTABILITY ANALYSIS; ASSOCIATIVE MEMORY; LOCAL STABILITY; GENERAL-CLASS; CONVERGENCE; MULTIPERIODICITY; DYNAMICS;
D O I：
10.1109/TNNLS.2021.3105519
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
The stability analysis of recurrent neural networks (RNNs) with multiple equilibria has received extensive interest since it is a prerequisite for successful applications of RNNs. With the increasing theoretical results on this topic, it is desirable to review the results for a systematical understanding of the state of the art. This article provides an overview of the stability results of RNNs with multiple equilibria including complete stability and multistability. First, preliminaries on the complete stability and multistability analysis of RNNs are introduced. Second, the complete stability results of RNNs are summarized. Third, the multistability results of various RNNs are reviewed in detail. Finally, future directions in these interesting topics are suggested.
引用
收藏
页码：1098 / 1111
页数：14
相关论文
共 159 条
[51]   Multistability of switched neural networks with sigmoidal activation functions under state-dependent switching [J].
Guo, Zhenyuan ;
Ou, Shiqin ;
Wang, Jun .
NEURAL NETWORKS, 2020, 122 :239-252
[52]   Multistability of Recurrent Neural Networks With Piecewise-Linear Radial Basis Functions and State-Dependent Switching Parameters [J].
Guo, Zhenyuan ;
Liu, Linlin ;
Wang, Jun .
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2020, 50 (11) :4458-4471
[53]   Multistability of Switched Neural Networks With Piecewise Linear Activation Functions Under State-Dependent Switching [J].
Guo, Zhenyuan ;
Liu, Linlin ;
Wang, Jun .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (07) :2052-2066
[54]  
Hens CR, 2012, PHYS REV E, V85, DOI 10.1103/PhysRevE.85.035202
[55]  
Hirose A, 2012, STUD COMPUT INTELL, V400, P1, DOI 10.1007/978-3-64227632-3
[56]   CONVERGENT ACTIVATION DYNAMICS IN CONTINUOUS-TIME NETWORKS [J].
HIRSCH, MW .
NEURAL NETWORKS, 1989, 2 (05) :331-349
[57]   NEURONS WITH GRADED RESPONSE HAVE COLLECTIVE COMPUTATIONAL PROPERTIES LIKE THOSE OF 2-STATE NEURONS [J].
HOPFIELD, JJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1984, 81 (10) :3088-3092
[58]   NEURAL NETWORKS AND PHYSICAL SYSTEMS WITH EMERGENT COLLECTIVE COMPUTATIONAL ABILITIES [J].
HOPFIELD, JJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1982, 79 (08) :2554-2558
[59]   Multistability of Delayed Hybrid Impulsive Neural Networks With Application to Associative Memories [J].
Hu, Bin ;
Guan, Zhi-Hong ;
Chen, Guanrong ;
Lewis, Frank L. .
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (05) :1537-1551
[60]   Global exponential periodicity and stability of discrete-time complex-valued recurrent neural networks with time-delays [J].
Hu, Jin ;
Wang, Jun .
NEURAL NETWORKS, 2015, 66 :119-130
12345678910