Distributed constrained optimization with periodic dynamic quantization

Jie Liu; Lulu Li; Daniel W.C. Ho

doi:10.1016/j.automatica.2023.111364

Distributed constrained optimization with periodic dynamic quantization

Jie Liu^*
, Lulu Li
, Daniel W.C. Ho

^*Corresponding author for this work

Department of Mathematics

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

6 Citations (Scopus)

Abstract

This paper uses the mirror descent algorithm with periodic dynamic quantization to solve constrained distributed optimization problems with limited communication channels. Due to the imperfect network environment, obtaining accurate information is impractical, and thus a communication scheme under quantization needs to be considered. A periodic dynamic quantizer with finite quantization levels is proposed in this paper to achieve exact optimization. Moreover, a time-varying control parameter in the mirror descent algorithm is designed to control the quantization error. After a comprehensive analysis, the proposed algorithm can obtain an optimal value, and the optimal convergence rate is O(1/T^0.25). © 2023 Elsevier Ltd

Original language	English
Article number	111364
Journal	Automatica
Volume	159
Online published	25 Oct 2023
DOIs	https://doi.org/10.1016/j.automatica.2023.111364
Publication status	Published - Jan 2024

Funding

This work was partially supported by Research Grants Council of the Hong Kong Special Administrative Region, China ( CityU 11202819 , CityU 11203521 , CityU 11213023 ). The material in this paper was not presented at any conference. This paper was recommended for publication in revised form by Associate Editor Sergio Grammatico under the direction of Editor Ian R. Petersen.

Research Keywords

Distributed optimization
Mirror descent algorithm
Periodic dynamic quantization

RGC Funding Information

RGC-funded

Access Output

10.1016/j.automatica.2023.111364

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{0990b7b76bba4f0bb3604e69ea405307,

title = "Distributed constrained optimization with periodic dynamic quantization",

abstract = "This paper uses the mirror descent algorithm with periodic dynamic quantization to solve constrained distributed optimization problems with limited communication channels. Due to the imperfect network environment, obtaining accurate information is impractical, and thus a communication scheme under quantization needs to be considered. A periodic dynamic quantizer with finite quantization levels is proposed in this paper to achieve exact optimization. Moreover, a time-varying control parameter in the mirror descent algorithm is designed to control the quantization error. After a comprehensive analysis, the proposed algorithm can obtain an optimal value, and the optimal convergence rate is O(1/T0.25). {\textcopyright} 2023 Elsevier Ltd",

keywords = "Distributed optimization, Mirror descent algorithm, Periodic dynamic quantization",

author = "Jie Liu and Lulu Li and Ho, \{Daniel W.C.\}",

year = "2024",

month = jan,

doi = "10.1016/j.automatica.2023.111364",

language = "English",

volume = "159",

journal = "Automatica",

issn = "0005-1098",

publisher = "Elsevier Ltd.",

}

TY - JOUR

T1 - Distributed constrained optimization with periodic dynamic quantization

AU - Liu, Jie

AU - Li, Lulu

AU - Ho, Daniel W.C.

PY - 2024/1

Y1 - 2024/1

N2 - This paper uses the mirror descent algorithm with periodic dynamic quantization to solve constrained distributed optimization problems with limited communication channels. Due to the imperfect network environment, obtaining accurate information is impractical, and thus a communication scheme under quantization needs to be considered. A periodic dynamic quantizer with finite quantization levels is proposed in this paper to achieve exact optimization. Moreover, a time-varying control parameter in the mirror descent algorithm is designed to control the quantization error. After a comprehensive analysis, the proposed algorithm can obtain an optimal value, and the optimal convergence rate is O(1/T0.25). © 2023 Elsevier Ltd

AB - This paper uses the mirror descent algorithm with periodic dynamic quantization to solve constrained distributed optimization problems with limited communication channels. Due to the imperfect network environment, obtaining accurate information is impractical, and thus a communication scheme under quantization needs to be considered. A periodic dynamic quantizer with finite quantization levels is proposed in this paper to achieve exact optimization. Moreover, a time-varying control parameter in the mirror descent algorithm is designed to control the quantization error. After a comprehensive analysis, the proposed algorithm can obtain an optimal value, and the optimal convergence rate is O(1/T0.25). © 2023 Elsevier Ltd

KW - Distributed optimization

KW - Mirror descent algorithm

KW - Periodic dynamic quantization

UR - https://www.scopus.com/pages/publications/85174612110

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85174612110&origin=recordpage

U2 - 10.1016/j.automatica.2023.111364

DO - 10.1016/j.automatica.2023.111364

M3 - RGC 21 - Publication in refereed journal

SN - 0005-1098

VL - 159

JO - Automatica

JF - Automatica

M1 - 111364

ER -

Distributed constrained optimization with periodic dynamic quantization

Abstract

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Distributed Mirror Descent Algorithm over Multi-agent Networks with Imperfect Communication

GRF: Distributed Optimization over Multi-agent Networks

GRF: Nonlinear Fusion Estimation for Networked Sensor Systems

Cite this

Distributed constrained optimization with periodic dynamic quantization

Abstract

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Distributed Mirror Descent Algorithm over Multi-agent Networks with Imperfect Communication

GRF: Distributed Optimization over Multi-agent Networks

GRF: Nonlinear Fusion Estimation for Networked Sensor Systems

Cite this