Predefined-time synchronization of competitive neural networks

Chuan Chen; Ling Mi; Zhongqiang Liu; Baolin Qiu; Hui Zhao; Lijuan Xu

doi:10.1016/j.neunet.2021.06.026

Predefined-time synchronization of competitive neural networks

Neural Netw. 2021 Oct:142:492-499. doi: 10.1016/j.neunet.2021.06.026. Epub 2021 Jul 7.

Authors

Chuan Chen¹, Ling Mi², Zhongqiang Liu³, Baolin Qiu⁴, Hui Zhao⁵, Lijuan Xu⁶

Affiliations

¹ School of Cyber Security, Qilu University of Technology (Shandong Academy of Sciences), Jinan 250353, China; State key Laboratory of Networking and Switching Technology (Beijing University of Posts and Telecommunications), Beijing 100876, China; Shandong Provincial Key Laboratory of Computer Networks, Shandong Computer Science Center(National Supercomputer Center in Jinan), Qilu University of Technology (Shandong Academy of Sciences), Jinan 250014, China. Electronic address: chenchuan2019@qlu.edu.cn.
² School of Mathematics and Statistics, Qilu University of Technology (Shandong Academy of Sciences), Jinan 250353, China. Electronic address: miling@lyu.edu.cn.
³ School of Mathematics and Information Science, Henan Polytechnic University, Jiaozuo 454003, China. Electronic address: zhongqiang@hpu.edu.cn.
⁴ School of Information Technology, Jiangxi University of Finance and Economics, Nanchang 330032, China. Electronic address: qiubaolin@jxufe.edu.cn.
⁵ Shandong Provincial Key Laboratory of Network Based Intelligent Computing, School of Information Science and Engineering, University of Jinan, Jinan 250022, China. Electronic address: ise_zhaohui@ujn.edu.cn.
⁶ Shandong Provincial Key Laboratory of Computer Networks, Shandong Computer Science Center(National Supercomputer Center in Jinan), Qilu University of Technology (Shandong Academy of Sciences), Jinan 250014, China. Electronic address: xulj@sdas.org.

PMID: 34280692
DOI: 10.1016/j.neunet.2021.06.026

Abstract

In this paper, the predefined-time synchronization of competitive neural networks (CNNs) is researched based on two different predefined-time stability theorems. In view of the bilayer structure of CNNs, we design two bilayer predefined-time controllers. The first controller utilizes sign function, while the second controller utilizes exponential function and Lyapunov function. In these two controllers, the predefined time is set as a controller parameter, and it can be an arbitrary positive constant. Under these two controllers, the considered CNNs can achieve synchronization within the predefined time regardless of the initial values. A specific example is presented to validate the theoretical results.

Keywords: Competitive neural networks; Predefined-time synchronization; The predefined time.

MeSH terms

Algorithms*
Neural Networks, Computer*
Time Factors