POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization

Qijing Wang; Martin D.F. Wong; Evangeline F.Y. Young

doi:10.1007/978-981-95-3001-4_32

POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization

Qijing Wang^*
, Martin D.F. Wong
, Evangeline F.Y. Young

^*Corresponding author for this work

Office of the Provost

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

Parameters play a key role in ensuring the expected behaviors of systems or achieving certain objectives, which gives rise to countless parameter optimization (PO) frameworks. In view of their shortcomings of weak adaptability to different scenarios caused by internal predefined configurations, this paper proposes a general controller named POAgent based on an efficient learning paradigm and multi-agent reinforcement learning, which can adaptively adjust the configurations and guide the PO process towards better outcomes according to the on-site situations. Experimental results show that significant improvements can be achieved when incorporating it into an existing SOTA PO framework.

Original language	English
Title of host publication	Knowledge Science, Engineering and Management
Subtitle of host publication	18th International Conference, KSEM 2025, Macao, China, August 4–7, 2025, Proceedings, Part I
Editors	Tianqing Zhu, Wanlei Zhou, Congcong Zhu
Place of Publication	Singapore
Publisher	Springer
Pages	430-442
Number of pages	13
ISBN (Electronic)	9789819530014
ISBN (Print)	9789819530007
DOIs	https://doi.org/10.1007/978-981-95-3001-4_32
Publication status	Published - 16 Nov 2025
Event	18th International Conference on Knowledge Science, Engineering and Management - Wynn Palace, Macao, China Duration: 4 Aug 2025 → 7 Aug 2025 https://ksem2025.scimeeting.cn/ (Conference website) https://ksem2025.scimeeting.cn/en/web/index/27434_2718139 (Conference program) https://link.springer.com/book/10.1007/978-981-95-3001-4 (Conference proceeding)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	15919
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Name	Lecture Notes in Artificial Intelligence
Publisher	Springer
ISSN (Print)	2945-9133
ISSN (Electronic)	2945-9141

Name	KSEM: International Conference on Knowledge Science, Engineering and Management
Publisher	Springer

Conference

Conference	18th International Conference on Knowledge Science, Engineering and Management
Abbreviated title	KSEM 2025
Country/Territory	China
City	Macao
Period	4/08/25 → 7/08/25
Internet address	https://ksem2025.scimeeting.cn/ (Conference website) https://ksem2025.scimeeting.cn/en/web/index/27434_2718139 (Conference program) https://link.springer.com/book/10.1007/978-981-95-3001-4 (Conference proceeding)

User-Defined Keywords

Adaptability
Machine learning
Parameter optimization

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-981-95-3001-4_32

Cite this

Wang, Q., Wong, M. D. F., & Young, E. F. Y. (2025). POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization. In T. Zhu, W. Zhou, & C. Zhu (Eds.), Knowledge Science, Engineering and Management: 18th International Conference, KSEM 2025, Macao, China, August 4–7, 2025, Proceedings, Part I (pp. 430-442). (Lecture Notes in Computer Science; Vol. 15919), (Lecture Notes in Artificial Intelligence), (KSEM: International Conference on Knowledge Science, Engineering and Management). Springer. https://doi.org/10.1007/978-981-95-3001-4_32

Wang, Qijing ; Wong, Martin D.F. ; Young, Evangeline F.Y. / POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization. Knowledge Science, Engineering and Management: 18th International Conference, KSEM 2025, Macao, China, August 4–7, 2025, Proceedings, Part I. editor / Tianqing Zhu ; Wanlei Zhou ; Congcong Zhu. Singapore : Springer, 2025. pp. 430-442 (Lecture Notes in Computer Science). (Lecture Notes in Artificial Intelligence). (KSEM: International Conference on Knowledge Science, Engineering and Management).

@inproceedings{5446d8820bd3455981ef6ac8622e369f,

title = "POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization",

abstract = "Parameters play a key role in ensuring the expected behaviors of systems or achieving certain objectives, which gives rise to countless parameter optimization (PO) frameworks. In view of their shortcomings of weak adaptability to different scenarios caused by internal predefined configurations, this paper proposes a general controller named POAgent based on an efficient learning paradigm and multi-agent reinforcement learning, which can adaptively adjust the configurations and guide the PO process towards better outcomes according to the on-site situations. Experimental results show that significant improvements can be achieved when incorporating it into an existing SOTA PO framework.",

keywords = "Adaptability, Machine learning, Parameter optimization",

author = "Qijing Wang and Wong, \{Martin D.F.\} and Young, \{Evangeline F.Y.\}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.; 18th International Conference on Knowledge Science, Engineering and Management, KSEM 2025 ; Conference date: 04-08-2025 Through 07-08-2025",

year = "2025",

month = nov,

day = "16",

doi = "10.1007/978-981-95-3001-4\_32",

language = "English",

isbn = "9789819530007",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "430--442",

editor = "Tianqing Zhu and Wanlei Zhou and Congcong Zhu",

booktitle = "Knowledge Science, Engineering and Management",

address = "Singapore",

url = "https://ksem2025.scimeeting.cn/, https://ksem2025.scimeeting.cn/en/web/index/27434\_2718139, https://link.springer.com/book/10.1007/978-981-95-3001-4",

}

Wang, Q, Wong, MDF & Young, EFY 2025, POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization. in T Zhu, W Zhou & C Zhu (eds), Knowledge Science, Engineering and Management: 18th International Conference, KSEM 2025, Macao, China, August 4–7, 2025, Proceedings, Part I. Lecture Notes in Computer Science, vol. 15919, Lecture Notes in Artificial Intelligence, KSEM: International Conference on Knowledge Science, Engineering and Management, Springer, Singapore, pp. 430-442, 18th International Conference on Knowledge Science, Engineering and Management, Macao, China, 4/08/25. https://doi.org/10.1007/978-981-95-3001-4_32

POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization. / Wang, Qijing; Wong, Martin D.F.; Young, Evangeline F.Y.
Knowledge Science, Engineering and Management: 18th International Conference, KSEM 2025, Macao, China, August 4–7, 2025, Proceedings, Part I. ed. / Tianqing Zhu; Wanlei Zhou; Congcong Zhu. Singapore: Springer, 2025. p. 430-442 (Lecture Notes in Computer Science; Vol. 15919), (Lecture Notes in Artificial Intelligence), (KSEM: International Conference on Knowledge Science, Engineering and Management).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization

AU - Wang, Qijing

AU - Wong, Martin D.F.

AU - Young, Evangeline F.Y.

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

PY - 2025/11/16

Y1 - 2025/11/16

N2 - Parameters play a key role in ensuring the expected behaviors of systems or achieving certain objectives, which gives rise to countless parameter optimization (PO) frameworks. In view of their shortcomings of weak adaptability to different scenarios caused by internal predefined configurations, this paper proposes a general controller named POAgent based on an efficient learning paradigm and multi-agent reinforcement learning, which can adaptively adjust the configurations and guide the PO process towards better outcomes according to the on-site situations. Experimental results show that significant improvements can be achieved when incorporating it into an existing SOTA PO framework.

AB - Parameters play a key role in ensuring the expected behaviors of systems or achieving certain objectives, which gives rise to countless parameter optimization (PO) frameworks. In view of their shortcomings of weak adaptability to different scenarios caused by internal predefined configurations, this paper proposes a general controller named POAgent based on an efficient learning paradigm and multi-agent reinforcement learning, which can adaptively adjust the configurations and guide the PO process towards better outcomes according to the on-site situations. Experimental results show that significant improvements can be achieved when incorporating it into an existing SOTA PO framework.

KW - Adaptability

KW - Machine learning

KW - Parameter optimization

UR - https://www.scopus.com/pages/publications/105022980089

UR - https://link.springer.com/chapter/10.1007/978-981-95-3001-4_32

U2 - 10.1007/978-981-95-3001-4_32

DO - 10.1007/978-981-95-3001-4_32

M3 - Conference proceeding

AN - SCOPUS:105022980089

SN - 9789819530007

T3 - Lecture Notes in Computer Science

SP - 430

EP - 442

BT - Knowledge Science, Engineering and Management

A2 - Zhu, Tianqing

A2 - Zhou, Wanlei

A2 - Zhu, Congcong

PB - Springer

CY - Singapore

T2 - 18th International Conference on Knowledge Science, Engineering and Management

Y2 - 4 August 2025 through 7 August 2025

ER -

Wang Q, Wong MDF, Young EFY. POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization. In Zhu T, Zhou W, Zhu C, editors, Knowledge Science, Engineering and Management: 18th International Conference, KSEM 2025, Macao, China, August 4–7, 2025, Proceedings, Part I. Singapore: Springer. 2025. p. 430-442. (Lecture Notes in Computer Science). (Lecture Notes in Artificial Intelligence). (KSEM: International Conference on Knowledge Science, Engineering and Management). doi: 10.1007/978-981-95-3001-4_32

POAgent: A Multi-agent Controller Towards Adaptive Parameter Optimization

Abstract

Publication series

Conference

User-Defined Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this