An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization

Daniel Herring; Michael Kirley; Xin Yao

doi:10.1145/3583133.3590624

An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization

Daniel Herring
, Michael Kirley
, Xin Yao

University of Melbourne
University of Birmingham

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding › peer-review

1 Citation (Scopus)

Abstract

Dynamic Multi-objective Optimization problems (DMOPs) can represent formulations of complex realistic scenarios in industrial, logistics and energy domains. Consistent and comparable testing on DMOP benchmarks has seen recent progress in tools for comprehensive and reproducible testing. Ranges of possible dynamic instances of DMOPs can be generated defined by their frequency and severity of change parameters. Here, a combination of machine learning and evolutionary algorithm techniques allow for an efficient determination of algorithm performance limits. An iterative Support Vector Machine model is integrated with the existing Dynamic Parameter Testing Platform (DPTP) for selective (rather than exhaustive) evaluation of dynamic instances to inform on algorithm capability.

Original language	English
Title of host publication	GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion
Publisher	Association for Computing Machinery, Inc
Pages	367-370
Number of pages	4
ISBN (Electronic)	9798400701207
DOIs	https://doi.org/10.1145/3583133.3590624
Publication status	Published - 15 Jul 2023
Externally published	Yes
Event	2023 Genetic and Evolutionary Computation Conference Companion, GECCO 2023 Companion - Lisbon, Portugal Duration: 15 Jul 2023 → 19 Jul 2023

Conference

Conference	2023 Genetic and Evolutionary Computation Conference Companion, GECCO 2023 Companion
Country/Territory	Portugal
City	Lisbon
Period	15/07/23 → 19/07/23

Bibliographical note

Publisher Copyright:
© 2023 Copyright held by the owner/author(s).

Keywords

dynamic multi-objective optimization
dynamic optimization
evolutionary algorithm
machine learning
support vector machine

ASJC Scopus subject areas

Software
Computational Theory and Mathematics
Computer Science Applications

Access to Document

10.1145/3583133.3590624

Cite this

An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization. / Herring, Daniel; Kirley, Michael; Yao, Xin.
GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion. Association for Computing Machinery, Inc, 2023. p. 367-370.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding › peer-review

Herring, D, Kirley, M & Yao, X 2023, An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization. in GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion. Association for Computing Machinery, Inc, pp. 367-370, 2023 Genetic and Evolutionary Computation Conference Companion, GECCO 2023 Companion, Lisbon, Portugal, 15/07/23. https://doi.org/10.1145/3583133.3590624

@inproceedings{7aab0ff2765a49a0a7b9eefb53d8e534,

title = "An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization",

abstract = "Dynamic Multi-objective Optimization problems (DMOPs) can represent formulations of complex realistic scenarios in industrial, logistics and energy domains. Consistent and comparable testing on DMOP benchmarks has seen recent progress in tools for comprehensive and reproducible testing. Ranges of possible dynamic instances of DMOPs can be generated defined by their frequency and severity of change parameters. Here, a combination of machine learning and evolutionary algorithm techniques allow for an efficient determination of algorithm performance limits. An iterative Support Vector Machine model is integrated with the existing Dynamic Parameter Testing Platform (DPTP) for selective (rather than exhaustive) evaluation of dynamic instances to inform on algorithm capability.",

keywords = "dynamic multi-objective optimization, dynamic optimization, evolutionary algorithm, machine learning, support vector machine",

author = "Daniel Herring and Michael Kirley and Xin Yao",

note = "Publisher Copyright: {\textcopyright} 2023 Copyright held by the owner/author(s).; 2023 Genetic and Evolutionary Computation Conference Companion, GECCO 2023 Companion ; Conference date: 15-07-2023 Through 19-07-2023",

year = "2023",

month = jul,

day = "15",

doi = "10.1145/3583133.3590624",

language = "English",

pages = "367--370",

booktitle = "GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion",

publisher = "Association for Computing Machinery, Inc",

address = "United States",

}

TY - GEN

T1 - An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization

AU - Herring, Daniel

AU - Kirley, Michael

AU - Yao, Xin

PY - 2023/7/15

Y1 - 2023/7/15

N2 - Dynamic Multi-objective Optimization problems (DMOPs) can represent formulations of complex realistic scenarios in industrial, logistics and energy domains. Consistent and comparable testing on DMOP benchmarks has seen recent progress in tools for comprehensive and reproducible testing. Ranges of possible dynamic instances of DMOPs can be generated defined by their frequency and severity of change parameters. Here, a combination of machine learning and evolutionary algorithm techniques allow for an efficient determination of algorithm performance limits. An iterative Support Vector Machine model is integrated with the existing Dynamic Parameter Testing Platform (DPTP) for selective (rather than exhaustive) evaluation of dynamic instances to inform on algorithm capability.

AB - Dynamic Multi-objective Optimization problems (DMOPs) can represent formulations of complex realistic scenarios in industrial, logistics and energy domains. Consistent and comparable testing on DMOP benchmarks has seen recent progress in tools for comprehensive and reproducible testing. Ranges of possible dynamic instances of DMOPs can be generated defined by their frequency and severity of change parameters. Here, a combination of machine learning and evolutionary algorithm techniques allow for an efficient determination of algorithm performance limits. An iterative Support Vector Machine model is integrated with the existing Dynamic Parameter Testing Platform (DPTP) for selective (rather than exhaustive) evaluation of dynamic instances to inform on algorithm capability.

KW - dynamic multi-objective optimization

KW - dynamic optimization

KW - evolutionary algorithm

KW - machine learning

KW - support vector machine

UR - https://www.scopus.com/pages/publications/85169007623

U2 - 10.1145/3583133.3590624

DO - 10.1145/3583133.3590624

M3 - Conference proceeding

AN - SCOPUS:85169007623

SP - 367

EP - 370

BT - GECCO 2023 Companion - Proceedings of the 2023 Genetic and Evolutionary Computation Conference Companion

PB - Association for Computing Machinery, Inc

T2 - 2023 Genetic and Evolutionary Computation Conference Companion, GECCO 2023 Companion

Y2 - 15 July 2023 through 19 July 2023

ER -

An Iterative Machine Learning Approach to Informative Performance Reporting in Dynamic Multi-Objective Optimization

Abstract

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this