Open AccessGlobal exponential stability analysis of anti-periodic of discontinuous BAM neural networks with time-varying delays
Author(s) -
N. Radhakrishnan,
R. Kodeeswaran,
R. Raja,
C. Maharajan,
A. Stephen
Publication year - 2021
Publication title -
journal of physics: conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
ISSN - 1742-6588
DOI - 10.1088/1742-6596/1850/1/012098
Subject(s) - bidirectional associative memory , exponential stability , differential inclusion , artificial neural network , computer science , control theory (sociology) , content addressable memory , stability (learning theory) , linear matrix inequality , mathematics , lyapunov function , nonlinear system , mathematical optimization , artificial intelligence , physics , control (management) , quantum mechanics , machine learning
Discontinuous system is playing an increasingly important role in terms of both theory and applications. In this paper, we are concerned with discontinuous BAM (bidirectional associative memory) neural networks with time-varying delays. Under the basic framework of Filippov solution, by means of differential inclusions theory, inequality technique, fundamental solution matrix of coefficients and the non-smooth analysis theory with Lyapunov-like approach, some new sufficient criteria are given to ascertain the existence and globally exponential stability of the anti-periodic solutions for the considered BAM neural networks. Simulation results of two topical numerical examples are exploited to illustrate the improvement and advantages of the established theoretical results in comparison with some existing results. Some previous known results are extended and complemented.