Stochastic stabilization of hybrid neural networks by periodically intermittent control based on discrete-time state observations

Wei Mao; Surong You; Yanan Jiang; Xuerong Mao

doi:10.1016/j.nahs.2023.101331

Stochastic stabilization of hybrid neural networks by periodically intermittent control based on discrete-time state observations

Wei Mao, Surong You, Yanan Jiang, Xuerong Mao

Mathematics And Statistics

Research output: Contribution to journal › Article › peer-review

14 Citations (Scopus)

8 Downloads (Pure)

Abstract

This paper is concerned with stabilization of hybrid neural networks by intermittent control based on continuous or discrete-time state observations. By means of exponential martingale inequality and the ergodic property of the Markov chain, we establish a sufficient stability criterion on hybrid neural networks by intermittent control based on continuous-time state observations. Meantime, by M-matrix theory and comparison method, we show that hybrid neural networks can be stabilized by intermittent control based on discrete-time state observations. Finally, two examples are presented to illustrate our theory.

Original language	English
Article number	101331
Number of pages	31
Journal	Nonlinear Analysis: Hybrid Systems
Volume	48
Early online date	14 Jan 2023
DOIs	https://doi.org/10.1016/j.nahs.2023.101331
Publication status	Published - 31 May 2023

Keywords

stochastic stabilization
hybrid stochastic neural networks
periodically intermittent control
discrete-time state observation

Access to Document

10.1016/j.nahs.2023.101331

Mao-etal-NAHS-2023-Stochastic-stabilization-of-hybrid-neural-networksAccepted author manuscript, 1 MBLicence: CC BY-NC-ND 4.0

Ergodicity and invariant measures of stochastic delay systems driven by various noises and their applications (Prof. Fuke Wu)
Mao, X. (Principal Investigator)
Royal Society
16/03/17 → 15/06/20
Project: Research Fellowship
Long-time dynamics of numerical solutions of stochastic differential equations
Mao, X. (Principal Investigator)
Royal Society
1/10/16 → 30/09/21
Project: Research
Numerical Analysis of Stochastic Differential Equations: New Challenges
Mao, X. (Principal Investigator)
Leverhulme Trust
1/10/15 → 30/09/17
Project: Research Fellowship

Cite this

@article{9edc4501b7b64a668fd3f21e8cd51d11,

title = "Stochastic stabilization of hybrid neural networks by periodically intermittent control based on discrete-time state observations",

abstract = "This paper is concerned with stabilization of hybrid neural networks by intermittent control based on continuous or discrete-time state observations. By means of exponential martingale inequality and the ergodic property of the Markov chain, we establish a sufficient stability criterion on hybrid neural networks by intermittent control based on continuous-time state observations. Meantime, by M-matrix theory and comparison method, we show that hybrid neural networks can be stabilized by intermittent control based on discrete-time state observations. Finally, two examples are presented to illustrate our theory.",

keywords = "stochastic stabilization, hybrid stochastic neural networks, periodically intermittent control, discrete-time state observation",

author = "Wei Mao and Surong You and Yanan Jiang and Xuerong Mao",

year = "2023",

month = may,

day = "31",

doi = "10.1016/j.nahs.2023.101331",

language = "English",

volume = "48",

journal = "Nonlinear Analysis: Hybrid Systems",

issn = "1751-570X",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Stochastic stabilization of hybrid neural networks by periodically intermittent control based on discrete-time state observations

AU - Mao, Wei

AU - You, Surong

AU - Jiang, Yanan

AU - Mao, Xuerong

PY - 2023/5/31

Y1 - 2023/5/31

N2 - This paper is concerned with stabilization of hybrid neural networks by intermittent control based on continuous or discrete-time state observations. By means of exponential martingale inequality and the ergodic property of the Markov chain, we establish a sufficient stability criterion on hybrid neural networks by intermittent control based on continuous-time state observations. Meantime, by M-matrix theory and comparison method, we show that hybrid neural networks can be stabilized by intermittent control based on discrete-time state observations. Finally, two examples are presented to illustrate our theory.

AB - This paper is concerned with stabilization of hybrid neural networks by intermittent control based on continuous or discrete-time state observations. By means of exponential martingale inequality and the ergodic property of the Markov chain, we establish a sufficient stability criterion on hybrid neural networks by intermittent control based on continuous-time state observations. Meantime, by M-matrix theory and comparison method, we show that hybrid neural networks can be stabilized by intermittent control based on discrete-time state observations. Finally, two examples are presented to illustrate our theory.

KW - stochastic stabilization

KW - hybrid stochastic neural networks

KW - periodically intermittent control

KW - discrete-time state observation

U2 - 10.1016/j.nahs.2023.101331

DO - 10.1016/j.nahs.2023.101331

M3 - Article

SN - 1751-570X

VL - 48

JO - Nonlinear Analysis: Hybrid Systems

JF - Nonlinear Analysis: Hybrid Systems

M1 - 101331

ER -

Stochastic stabilization of hybrid neural networks by periodically intermittent control based on discrete-time state observations

Abstract

Keywords

Access to Document

Fingerprint

Projects

Ergodicity and invariant measures of stochastic delay systems driven by various noises and their applications (Prof. Fuke Wu)

Long-time dynamics of numerical solutions of stochastic differential equations

Numerical Analysis of Stochastic Differential Equations: New Challenges

Cite this