A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization

Yangxia Li; Zicong Xia; Yang Liu; Jun Wang

doi:10.1016/j.neunet.2024.106525

A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization

Yangxia Li
, Zicong Xia
, Yang Liu^*
, Jun Wang^*

^*Corresponding author for this work

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

11 Citations (Scopus)

Abstract

In this paper, two two-timescale projection neural networks are proposed based on the majorization-minimization principle for nonconvex optimization and distributed nonconvex optimization. They are proved to be globally convergent to Karush–Kuhn–Tucker points. A collaborative neurodynamic approach leverages multiple two-timescale projection neural networks repeatedly re-initialized using a meta-heuristic rule for global optimization and distributed global optimization. Two numerical examples are elaborated to demonstrate the efficacy of the proposed approaches. © 2024 Elsevier Ltd.

Original language	English
Article number	106525
Journal	Neural Networks
Volume	179
Online published	11 Jul 2024
DOIs	https://doi.org/10.1016/j.neunet.2024.106525
Publication status	Published - Nov 2024

Funding

This work was partially supported by the National Natural Science Foundation of China (62173308, 623B2018), the Jinhua Science and Technology Project, China (2022-1-042), the Hong Kong Research Grants Council, Hong Kong (11203721).

Research Keywords

Collaborative neurodynamic optimization
Distributed optimization
Global optimization
Majorization-minimization principle
Projection neural network

RGC Funding Information

RGC-funded

Access Output

10.1016/j.neunet.2024.106525

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{4eac14b7f7dd412f8d8e01290ed90f4f,

title = "A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization",

abstract = "In this paper, two two-timescale projection neural networks are proposed based on the majorization-minimization principle for nonconvex optimization and distributed nonconvex optimization. They are proved to be globally convergent to Karush–Kuhn–Tucker points. A collaborative neurodynamic approach leverages multiple two-timescale projection neural networks repeatedly re-initialized using a meta-heuristic rule for global optimization and distributed global optimization. Two numerical examples are elaborated to demonstrate the efficacy of the proposed approaches. {\textcopyright} 2024 Elsevier Ltd.",

keywords = "Collaborative neurodynamic optimization, Distributed optimization, Global optimization, Majorization-minimization principle, Projection neural network",

author = "Yangxia Li and Zicong Xia and Yang Liu and Jun Wang",

year = "2024",

month = nov,

doi = "10.1016/j.neunet.2024.106525",

language = "English",

volume = "179",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Ltd.",

}

A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization. / Li, Yangxia; Xia, Zicong; Liu, Yang et al.
In: Neural Networks, Vol. 179, 106525, 11.2024.

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

TY - JOUR

T1 - A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization

AU - Li, Yangxia

AU - Xia, Zicong

AU - Liu, Yang

AU - Wang, Jun

PY - 2024/11

Y1 - 2024/11

N2 - In this paper, two two-timescale projection neural networks are proposed based on the majorization-minimization principle for nonconvex optimization and distributed nonconvex optimization. They are proved to be globally convergent to Karush–Kuhn–Tucker points. A collaborative neurodynamic approach leverages multiple two-timescale projection neural networks repeatedly re-initialized using a meta-heuristic rule for global optimization and distributed global optimization. Two numerical examples are elaborated to demonstrate the efficacy of the proposed approaches. © 2024 Elsevier Ltd.

AB - In this paper, two two-timescale projection neural networks are proposed based on the majorization-minimization principle for nonconvex optimization and distributed nonconvex optimization. They are proved to be globally convergent to Karush–Kuhn–Tucker points. A collaborative neurodynamic approach leverages multiple two-timescale projection neural networks repeatedly re-initialized using a meta-heuristic rule for global optimization and distributed global optimization. Two numerical examples are elaborated to demonstrate the efficacy of the proposed approaches. © 2024 Elsevier Ltd.

KW - Collaborative neurodynamic optimization

KW - Distributed optimization

KW - Global optimization

KW - Majorization-minimization principle

KW - Projection neural network

UR - https://www.scopus.com/pages/publications/85199077487

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85199077487&origin=recordpage

U2 - 10.1016/j.neunet.2024.106525

DO - 10.1016/j.neunet.2024.106525

M3 - RGC 21 - Publication in refereed journal

SN - 0893-6080

VL - 179

JO - Neural Networks

JF - Neural Networks

M1 - 106525

ER -

A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization

Abstract

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Neurodynamics-driven Optimization and Control of Intelligent Heating, Ventilation and Air Conditioning Systems

Cite this

A collaborative neurodynamic approach with two-timescale projection neural networks designed via majorization-minimization for global optimization and distributed global optimization

Abstract

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Neurodynamics-driven Optimization and Control of Intelligent Heating, Ventilation and Air Conditioning Systems

Cite this