Multistability of Hopfield neural networks with a designed discontinuous sawtooth-type activation function

被引:18
|
作者
Liu, Yang [1 ]
Huang, Xia [2 ]
Li, Yuxia [2 ]
Shen, Hao [3 ]
机构
[1] Shandong Univ Sci & Technol, Coll Math & Syst Sci, Qingdao 266590, Peoples R China
[2] Shandong Univ Sci & Technol, Coll Elect Engn & Automat, Qingdao 266590, Peoples R China
[3] Anhui Univ Technol, Elect & Informat Engn, Maanshan 243002, Peoples R China
基金
中国国家自然科学基金;
关键词
Discontinuous activation functions; Equilibrium point; Hopfield neural networks; Multistability; DYNAMICAL BEHAVIORS; ASSOCIATIVE MEMORY; MULTIPERIODICITY; STABILITY;
D O I
10.1016/j.neucom.2021.05.045
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper, a class of discontinuous sawtooth-type activation function is designed and the multistability of Hopfield neural networks (HNNs) with such kind of activation function is studied. By virtue of the Brouwer's fixed point theorem and the property of strictly diagonally dominant matrix (SDDM), some sufficient conditions are presented to ensure that the n-neuron HNN can have at least 7(n) equilibria, among which 4(n) equilibria are locally exponentially stable and the remaining 7(n)-4(n) equilibria are unstable. Then, the obtained results are extended to a more general case. We continue to increase the number of the peaks of the sawtooth-type activation function and we find that the n-neuron HNN can have (2k + 3)(n) equilibria, (k + 2)(n) of them are locally exponentially stable and the remaining equilibria are unstable. Therein, k denotes the total number of the peaks in the designed activation function. That is to say, there is a quantitative relationship between the number of the peaks and the number of the equilibria. It implies that one can improve the storage capacity of a HNN by increasing the number of the peaks of the activation function in theory and in practice. To some extent, this method is convenient and flexible. Compared with the existing results, HNN with the designed sawtooth-type activation function can have more total equilibria as well as more locally stable equilibria. Finally, two examples are presented to demonstrate the validity of the obtained results. (C) 2021 Elsevier B.V. All rights reserved.
引用
收藏
页码:189 / 201
页数:13
相关论文
共 50 条
  • [1] Multistability of neural networks with discontinuous activation function
    Huang, Gan
    Cao, Jinde
    COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION, 2008, 13 (10) : 2279 - 2289
  • [2] On Multistability of Competitive Neural Networks with Discontinuous Activation Functions
    Nie, Xiaobing
    Zheng, Wei Xing
    2014 4TH AUSTRALIAN CONTROL CONFERENCE (AUCC), 2014, : 245 - 250
  • [3] Multistability analysis of state-dependent switched Hopfield neural networks with the Gaussian-wavelet-type activation function
    Liu, Yang
    Wang, Zhen
    Huang, Xia
    MATHEMATICS AND COMPUTERS IN SIMULATION, 2022, 196 : 232 - 250
  • [4] Multistability in a class of stochastic Hopfield neural networks
    Chen, Wu-hua
    Luo, Shixian
    Lu, Xiaomei
    2014 33RD CHINESE CONTROL CONFERENCE (CCC), 2014, : 5259 - 5264
  • [5] Multistability of complex-valued neural networks with discontinuous activation functions
    Liang, Jinling
    Gong, Weiqiang
    Huang, Tingwen
    NEURAL NETWORKS, 2016, 84 : 125 - 142
  • [6] Multistability in a class of stochastic delayed Hopfield neural networks
    Chen, Wu-Hua
    Luo, Shixian
    Lu, Xiaomei
    NEURAL NETWORKS, 2015, 68 : 52 - 61
  • [7] Multistability and Instability of Neural Networks With Discontinuous Nonmonotonic Piecewise Linear Activation Functions
    Nie, Xiaobing
    Zheng, Wei Xing
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2015, 26 (11) : 2901 - 2913
  • [8] Multistability of Delayed Neural Networks with Discontinuous Activations
    Chen, Xiaofeng
    Zhou, Yafei
    Song, Qiankun
    ADVANCES IN NEURAL NETWORKS - ISNN 2010, PT 1, PROCEEDINGS, 2010, 6063 : 603 - +
  • [9] Multistability in impulsive hybrid Hopfield neural networks with distributed delays
    Kaslik, Eva
    Sivasundaram, Seenith
    NONLINEAR ANALYSIS-REAL WORLD APPLICATIONS, 2011, 12 (03) : 1640 - 1649
  • [10] Multistability of Fractional-Order Recurrent Neural Networks With Discontinuous and Nonmonotonic Activation Functions
    Huang, Yujiao
    Yuan, Xiaoyan
    Long, Haixia
    Fan, Xinggang
    Cai, Tiaoyang
    IEEE ACCESS, 2019, 7 : 116430 - 116437