A Trust-Region Projection Neural Network for Nonlinear Programming

Haoen Huang; Zhigang Zeng; Jun Wang

doi:10.1109/TNNLS.2025.3576600

A Trust-Region Projection Neural Network for Nonlinear Programming

Haoen Huang
, Zhigang Zeng^*
, Jun Wang^*

^*Corresponding author for this work

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

4 Citations (Scopus)

Abstract

The trust-region method and projection neural networks are two branches of optimization approaches with different operational principles and characteristics. In this article, a trust-region projection neural network (TRPNN) is proposed by integrating the trust-region method and projection neural networks. TRPNN is a discrete-time neurodynamic optimization model that inherits the exploration–exploitation capability of the trust-region method and the local search capability of projection neural networks. TRPNN is theoretically proven to be convergent to a Karush–Kuhn–Tuchker (KKT) point of nonlinear programming problems. The efficacy of TRPNNs leveraged in a collaborative neurodynamic framework is numerically demonstrated for global optimization in the presence of nonconvexity in objective functions or constraints. © 2025 IEEE.

Original language	English
Pages (from-to)	18737-18749
Journal	IEEE Transactions on Neural Networks and Learning Systems
Volume	36
Issue number	10
Online published	12 Jun 2025
DOIs	https://doi.org/10.1109/TNNLS.2025.3576600
Publication status	Published - Oct 2025

Funding

This work was supported in part by the National Key Research and Development Program of China under Grant 2021ZD0201300, in part by the Foundation for Outstanding Research Groups of Hubei Province of China under Grant 2025AFA012, in part by the 111 Project on Computational Intelligence and Intelligent Control under Grant B18024, and in part by the Research Grants Council of Hong Kong Special Administrative Region of China under Grant AoE/E-407/24-N.

Research Keywords

Nonlinear programming
projection neural network
trust-region method

RGC Funding Information

RGC-funded

Access Output

10.1109/TNNLS.2025.3576600

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{4bf37316e07c4de69e9edb95780cc7c4,

title = "A Trust-Region Projection Neural Network for Nonlinear Programming",

abstract = "The trust-region method and projection neural networks are two branches of optimization approaches with different operational principles and characteristics. In this article, a trust-region projection neural network (TRPNN) is proposed by integrating the trust-region method and projection neural networks. TRPNN is a discrete-time neurodynamic optimization model that inherits the exploration–exploitation capability of the trust-region method and the local search capability of projection neural networks. TRPNN is theoretically proven to be convergent to a Karush–Kuhn–Tuchker (KKT) point of nonlinear programming problems. The efficacy of TRPNNs leveraged in a collaborative neurodynamic framework is numerically demonstrated for global optimization in the presence of nonconvexity in objective functions or constraints. {\textcopyright} 2025 IEEE.",

keywords = "Nonlinear programming, projection neural network, trust-region method",

author = "Haoen Huang and Zhigang Zeng and Jun Wang",

year = "2025",

month = oct,

doi = "10.1109/TNNLS.2025.3576600",

language = "English",

volume = "36",

pages = "18737--18749",

journal = "IEEE Transactions on Neural Networks and Learning Systems",

issn = "2162-237X",

publisher = "IEEE Computational Intelligence Society",

number = "10",

}

TY - JOUR

T1 - A Trust-Region Projection Neural Network for Nonlinear Programming

AU - Huang, Haoen

AU - Zeng, Zhigang

AU - Wang, Jun

PY - 2025/10

Y1 - 2025/10

N2 - The trust-region method and projection neural networks are two branches of optimization approaches with different operational principles and characteristics. In this article, a trust-region projection neural network (TRPNN) is proposed by integrating the trust-region method and projection neural networks. TRPNN is a discrete-time neurodynamic optimization model that inherits the exploration–exploitation capability of the trust-region method and the local search capability of projection neural networks. TRPNN is theoretically proven to be convergent to a Karush–Kuhn–Tuchker (KKT) point of nonlinear programming problems. The efficacy of TRPNNs leveraged in a collaborative neurodynamic framework is numerically demonstrated for global optimization in the presence of nonconvexity in objective functions or constraints. © 2025 IEEE.

AB - The trust-region method and projection neural networks are two branches of optimization approaches with different operational principles and characteristics. In this article, a trust-region projection neural network (TRPNN) is proposed by integrating the trust-region method and projection neural networks. TRPNN is a discrete-time neurodynamic optimization model that inherits the exploration–exploitation capability of the trust-region method and the local search capability of projection neural networks. TRPNN is theoretically proven to be convergent to a Karush–Kuhn–Tuchker (KKT) point of nonlinear programming problems. The efficacy of TRPNNs leveraged in a collaborative neurodynamic framework is numerically demonstrated for global optimization in the presence of nonconvexity in objective functions or constraints. © 2025 IEEE.

KW - Nonlinear programming

KW - projection neural network

KW - trust-region method

UR - https://www.scopus.com/pages/publications/105008204984

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-105008204984&origin=recordpage

U2 - 10.1109/TNNLS.2025.3576600

DO - 10.1109/TNNLS.2025.3576600

M3 - RGC 21 - Publication in refereed journal

SN - 2162-237X

VL - 36

SP - 18737

EP - 18749

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

IS - 10

ER -

A Trust-Region Projection Neural Network for Nonlinear Programming

Abstract

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

AoE(UGC)-ExtU-Lead: AI-Powered Surgical Robots

Cite this

A Trust-Region Projection Neural Network for Nonlinear Programming

Abstract

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

AoE(UGC)-ExtU-Lead: AI-Powered Surgical Robots

Cite this