An improved stability criterion of neural networks with time-varying delays in the form of quadratic function using novel geometry-based conditions

Tae H. Lee; Myeong Jin Park; Ju H. Park

doi:10.1016/j.amc.2021.126226

An improved stability criterion of neural networks with time-varying delays in the form of quadratic function using novel geometry-based conditions

Tae H. Lee^*
, Myeong Jin Park
, Ju H. Park

^*Corresponding author for this work

Division of Electronic Engineering

Research output: Contribution to journal › Journal article › peer-review

32 Scopus citations

Abstract

In this paper, the stability problem of neural networks is addressed by considering time-varying delays. By proposing novel geometry-based negative conditions for the form of quadratic function and constructing new augmented Lyapunov-Krasovskii functionals, a novel stability criterion is derived. Finally, to show the effectiveness of the proposed criterion, several numerical examples are given.

Original language	English
Article number	126226
Journal	Applied Mathematics and Computation
Volume	404
DOIs	https://doi.org/10.1016/j.amc.2021.126226
State	Published - 2021.09.1

Keywords

Neural networks
Quadratic function
Stability
Time-varying delay

Quacquarelli Symonds(QS) Subject Topics

Mathematics

Access to Document

10.1016/j.amc.2021.126226

Cite this

@article{86e649ccbc2a4e398325b785962dc9a1,

title = "An improved stability criterion of neural networks with time-varying delays in the form of quadratic function using novel geometry-based conditions",

abstract = "In this paper, the stability problem of neural networks is addressed by considering time-varying delays. By proposing novel geometry-based negative conditions for the form of quadratic function and constructing new augmented Lyapunov-Krasovskii functionals, a novel stability criterion is derived. Finally, to show the effectiveness of the proposed criterion, several numerical examples are given.",

keywords = "Neural networks, Quadratic function, Stability, Time-varying delay",

author = "Lee, \{Tae H.\} and Park, \{Myeong Jin\} and Park, \{Ju H.\}",

note = "Publisher Copyright: {\textcopyright} 2021",

year = "2021",

month = sep,

day = "1",

doi = "10.1016/j.amc.2021.126226",

language = "English",

volume = "404",

journal = "Applied Mathematics and Computation",

issn = "0096-3003",

}

TY - JOUR

T1 - An improved stability criterion of neural networks with time-varying delays in the form of quadratic function using novel geometry-based conditions

AU - Lee, Tae H.

AU - Park, Myeong Jin

AU - Park, Ju H.

PY - 2021/9/1

Y1 - 2021/9/1

N2 - In this paper, the stability problem of neural networks is addressed by considering time-varying delays. By proposing novel geometry-based negative conditions for the form of quadratic function and constructing new augmented Lyapunov-Krasovskii functionals, a novel stability criterion is derived. Finally, to show the effectiveness of the proposed criterion, several numerical examples are given.

AB - In this paper, the stability problem of neural networks is addressed by considering time-varying delays. By proposing novel geometry-based negative conditions for the form of quadratic function and constructing new augmented Lyapunov-Krasovskii functionals, a novel stability criterion is derived. Finally, to show the effectiveness of the proposed criterion, several numerical examples are given.

KW - Neural networks

KW - Quadratic function

KW - Stability

KW - Time-varying delay

UR - https://www.scopus.com/pages/publications/85103691991

U2 - 10.1016/j.amc.2021.126226

DO - 10.1016/j.amc.2021.126226

M3 - Journal article

AN - SCOPUS:85103691991

SN - 0096-3003

VL - 404

JO - Applied Mathematics and Computation

JF - Applied Mathematics and Computation

M1 - 126226

ER -

An improved stability criterion of neural networks with time-varying delays in the form of quadratic function using novel geometry-based conditions

Abstract

Keywords

Quacquarelli Symonds(QS) Subject Topics

Access to Document

Other files and links

Fingerprint

Cite this