On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems

Oliver Ludger Preuß; Jeroen Rook; Heike Trautmann

doi:10.1007/978-3-031-56852-7_20

On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems

Oliver Ludger Preuß^*
, Jeroen Rook
, Heike Trautmann

^*Corresponding author for this work

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

5 Citations (Scopus)

185 Downloads (Pure)

Abstract

The complexity of Multi-Objective (MO) continuous optimisation problems arises from a combination of different characteristics, such as the level of multi-modality. Earlier studies revealed that there is a conflict between solver convergence in objective space and solution set diversity in the decision space, which is especially important in the multi-modal setting. We build on top of this observation and investigate this trade-off in a multi-objective manner by using multi-objective automated algorithm configuration (MO-AAC) on evolutionary multi-objective algorithms (EMOA). Our results show that MO-AAC is able to find configurations that outperform the default configuration as well as configurations found by single-objective AAC in regards to objective space convergence and diversity in decision space, leading to new recommendations for high-performing default settings.

Original language	English
Title of host publication	Applications of Evolutionary Computation
Subtitle of host publication	27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I
Editors	Stephen Smith, João Correia, Christian Cintrano
Pages	305-321
Number of pages	17
ISBN (Electronic)	978-3-031-56852-7
DOIs	https://doi.org/10.1007/978-3-031-56852-7_20
Publication status	Published - 21 Mar 2024
Event	27th International Conference on Applications of Evolutionary Computation, EvoApplications 2024 - Aberystwyth, United Kingdom Duration: 3 Apr 2024 → 5 Apr 2024 Conference number: 27

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer, Cham.
Volume	14634

Conference

Conference	27th International Conference on Applications of Evolutionary Computation, EvoApplications 2024
Abbreviated title	EvoApplications 2024
Country/Territory	United Kingdom
City	Aberystwyth
Period	3/04/24 → 5/04/24

Keywords

2024 OA procedure

Access to Document

10.1007/978-3-031-56852-7_20

978-3-031-56852-7_20Final published version, 1.02 MBLicence: Taverne

Cite this

Preuß, O. L., Rook, J., & Trautmann, H. (2024). On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems. In S. Smith, J. Correia, & C. Cintrano (Eds.), Applications of Evolutionary Computation: 27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I (pp. 305-321). (Lecture Notes in Computer Science; Vol. 14634). https://doi.org/10.1007/978-3-031-56852-7_20

Preuß, Oliver Ludger ; Rook, Jeroen ; Trautmann, Heike. / On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems. Applications of Evolutionary Computation: 27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I. editor / Stephen Smith ; João Correia ; Christian Cintrano. 2024. pp. 305-321 (Lecture Notes in Computer Science).

@inbook{6acabef15dfe43c6a0c97fde9c62240a,

title = "On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems",

abstract = "The complexity of Multi-Objective (MO) continuous optimisation problems arises from a combination of different characteristics, such as the level of multi-modality. Earlier studies revealed that there is a conflict between solver convergence in objective space and solution set diversity in the decision space, which is especially important in the multi-modal setting. We build on top of this observation and investigate this trade-off in a multi-objective manner by using multi-objective automated algorithm configuration (MO-AAC) on evolutionary multi-objective algorithms (EMOA). Our results show that MO-AAC is able to find configurations that outperform the default configuration as well as configurations found by single-objective AAC in regards to objective space convergence and diversity in decision space, leading to new recommendations for high-performing default settings.",

keywords = "2024 OA procedure",

author = "Preu{\ss}, \{Oliver Ludger\} and Jeroen Rook and Heike Trautmann",

year = "2024",

month = mar,

day = "21",

doi = "10.1007/978-3-031-56852-7\_20",

language = "English",

isbn = "978-3-031-56851-0",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Cham.",

pages = "305--321",

editor = "Stephen Smith and Jo{\~a}o Correia and Christian Cintrano",

booktitle = "Applications of Evolutionary Computation",

note = "27th International Conference on Applications of Evolutionary Computation, EvoApplications 2024, EvoApplications 2024 ; Conference date: 03-04-2024 Through 05-04-2024",

}

Preuß, OL, Rook, J & Trautmann, H 2024, On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems. in S Smith, J Correia & C Cintrano (eds), Applications of Evolutionary Computation: 27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I. Lecture Notes in Computer Science, vol. 14634, pp. 305-321, 27th International Conference on Applications of Evolutionary Computation, EvoApplications 2024, Aberystwyth, United Kingdom, 3/04/24. https://doi.org/10.1007/978-3-031-56852-7_20

On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems. / Preuß, Oliver Ludger; Rook, Jeroen; Trautmann, Heike.
Applications of Evolutionary Computation: 27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I. ed. / Stephen Smith; João Correia; Christian Cintrano. 2024. p. 305-321 (Lecture Notes in Computer Science; Vol. 14634).

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

TY - CHAP

T1 - On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems

AU - Preuß, Oliver Ludger

AU - Rook, Jeroen

AU - Trautmann, Heike

N1 - Conference code: 27

PY - 2024/3/21

Y1 - 2024/3/21

N2 - The complexity of Multi-Objective (MO) continuous optimisation problems arises from a combination of different characteristics, such as the level of multi-modality. Earlier studies revealed that there is a conflict between solver convergence in objective space and solution set diversity in the decision space, which is especially important in the multi-modal setting. We build on top of this observation and investigate this trade-off in a multi-objective manner by using multi-objective automated algorithm configuration (MO-AAC) on evolutionary multi-objective algorithms (EMOA). Our results show that MO-AAC is able to find configurations that outperform the default configuration as well as configurations found by single-objective AAC in regards to objective space convergence and diversity in decision space, leading to new recommendations for high-performing default settings.

AB - The complexity of Multi-Objective (MO) continuous optimisation problems arises from a combination of different characteristics, such as the level of multi-modality. Earlier studies revealed that there is a conflict between solver convergence in objective space and solution set diversity in the decision space, which is especially important in the multi-modal setting. We build on top of this observation and investigate this trade-off in a multi-objective manner by using multi-objective automated algorithm configuration (MO-AAC) on evolutionary multi-objective algorithms (EMOA). Our results show that MO-AAC is able to find configurations that outperform the default configuration as well as configurations found by single-objective AAC in regards to objective space convergence and diversity in decision space, leading to new recommendations for high-performing default settings.

KW - 2024 OA procedure

U2 - 10.1007/978-3-031-56852-7_20

DO - 10.1007/978-3-031-56852-7_20

M3 - Chapter

SN - 978-3-031-56851-0

T3 - Lecture Notes in Computer Science

SP - 305

EP - 321

BT - Applications of Evolutionary Computation

A2 - Smith, Stephen

A2 - Correia, João

A2 - Cintrano, Christian

T2 - 27th International Conference on Applications of Evolutionary Computation, EvoApplications 2024

Y2 - 3 April 2024 through 5 April 2024

ER -

Preuß OL, Rook J, Trautmann H. On the Potential of Multi-objective Automated Algorithm Configuration on Multi-modal Multi-objective Optimisation Problems. In Smith S, Correia J, Cintrano C, editors, Applications of Evolutionary Computation: 27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I. 2024. p. 305-321. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-56852-7_20