LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection

Cheng Gong; Ping Guo; Tianye Shu; Qingfu Zhang; Hisao Ishibuchi

doi:10.1007/978-3-031-70085-9_3

LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection

Cheng Gong
, Ping Guo
, Tianye Shu
, Qingfu Zhang
, Hisao Ishibuchi

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 12 - Chapter in an edited book (Author) › peer-review

1 Citation (Scopus)

Abstract

Hypervolume subset selection (HSS) plays an important role in various aspects of the field of evolutionary multi-objective optimization, such as environmental selection and post-processing for decision-making. The goal of these problems is to find the optimal subset that maximizes the hypervolume from a given candidate solution set. Many methods have been developed to solve or approximately solve different types of HSS problems. However, existing approaches cannot effectively solve HSS problems with a large number of objectives within a short computation time. This drawback directly limits their applicability as a component for developing new EMO algorithms. In this paper, we propose a novel learning-to-rank based framework, named LTR-HSS, for solving the challenging HSS problems with a large number of objectives. The experimental results show that, compared to other state-of-the-art HSS methods, our proposed LTR-HSS requires a shorter computation time to solve HSS problems with large numbers of objectives while achieving superior or competitive hypervolume performance. This demonstrates the potential of our method to be integrated into algorithms for many-objective optimization.
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG

Original language	English
Title of host publication	Parallel Problem Solving from Nature
Subtitle of host publication	PPSN XVII
Editors	Michael Affenzeller, Stephan M. Winkler, Anna V. Kononova, Heike Trautmann, Tea Tušar, Penousal Machado, Thomas Bäck
Publisher	Springer, Cham
Chapter	3
Pages	36-51
Number of pages	16
Edition	1
ISBN (Electronic)	9783031700859
ISBN (Print)	9783031700842
DOIs	https://doi.org/10.1007/978-3-031-70085-9_3
Publication status	Published - 7 Sept 2024

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	15151
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Funding

This work was supported by the National Key R&D Program of China (Grant No. 2023YFE0106300), National Natural Science Foundation of China (Grant No. 62250710163, 62376115, 62276223), Guangdong Provincial Key Laboratory (Grant No. 2020B121201001), and the Research Grants Council of the Hong Kong Special Administrative Region, China [GRF Project No. CityU 11215622].

Research Keywords

Hypervolume subset selection
Multi-objective optimization
Machine learning

RGC Funding Information

RGC-funded

Access Output

10.1007/978-3-031-70085-9_3

https://link.springer.com/10.1007/978-3-031-70085-9_3

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

GRF: Few for Many: A Non-Pareto Approach for Many Objective Optimization
ZHANG, Q. (Principal Investigator / Project Coordinator)
1/01/23 → …
Project: Research

Cite this

Gong, C., Guo, P., Shu, T., Zhang, Q., & Ishibuchi, H. (2024). LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection. In M. Affenzeller, S. M. Winkler, A. V. Kononova, H. Trautmann, T. Tušar, P. Machado, & T. Bäck (Eds.), Parallel Problem Solving from Nature: PPSN XVII (1 ed., pp. 36-51). (Lecture Notes in Computer Science; Vol. 15151). Springer, Cham. https://doi.org/10.1007/978-3-031-70085-9_3

Gong, Cheng ; Guo, Ping ; Shu, Tianye et al. / LTR-HSS : A Learning-to-Rank Based Framework for Hypervolume Subset Selection. Parallel Problem Solving from Nature: PPSN XVII. editor / Michael Affenzeller ; Stephan M. Winkler ; Anna V. Kononova ; Heike Trautmann ; Tea Tušar ; Penousal Machado ; Thomas Bäck. 1. ed. Springer, Cham, 2024. pp. 36-51 (Lecture Notes in Computer Science).

@inbook{f13ee273425348dc9c3cfacdedc2debc,

title = "LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection",

abstract = "Hypervolume subset selection (HSS) plays an important role in various aspects of the field of evolutionary multi-objective optimization, such as environmental selection and post-processing for decision-making. The goal of these problems is to find the optimal subset that maximizes the hypervolume from a given candidate solution set. Many methods have been developed to solve or approximately solve different types of HSS problems. However, existing approaches cannot effectively solve HSS problems with a large number of objectives within a short computation time. This drawback directly limits their applicability as a component for developing new EMO algorithms. In this paper, we propose a novel learning-to-rank based framework, named LTR-HSS, for solving the challenging HSS problems with a large number of objectives. The experimental results show that, compared to other state-of-the-art HSS methods, our proposed LTR-HSS requires a shorter computation time to solve HSS problems with large numbers of objectives while achieving superior or competitive hypervolume performance. This demonstrates the potential of our method to be integrated into algorithms for many-objective optimization. {\textcopyright} 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG",

keywords = "Hypervolume subset selection, Multi-objective optimization, Machine learning",

author = "Cheng Gong and Ping Guo and Tianye Shu and Qingfu Zhang and Hisao Ishibuchi",

year = "2024",

month = sep,

day = "7",

doi = "10.1007/978-3-031-70085-9\_3",

language = "English",

isbn = "9783031700842",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Cham",

pages = "36--51",

editor = "Michael Affenzeller and Winkler, \{Stephan M.\} and Kononova, \{Anna V.\} and Heike Trautmann and Tea Tu{\v s}ar and Penousal Machado and Thomas B{\"a}ck",

booktitle = "Parallel Problem Solving from Nature",

edition = "1",

}

Gong, C , Guo, P, Shu, T, Zhang, Q & Ishibuchi, H 2024, LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection. in M Affenzeller, SM Winkler, AV Kononova, H Trautmann, T Tušar, P Machado & T Bäck (eds), Parallel Problem Solving from Nature: PPSN XVII. 1 edn, Lecture Notes in Computer Science, vol. 15151, Springer, Cham, pp. 36-51. https://doi.org/10.1007/978-3-031-70085-9_3

LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection. / Gong, Cheng ; Guo, Ping; Shu, Tianye et al.
Parallel Problem Solving from Nature: PPSN XVII. ed. / Michael Affenzeller; Stephan M. Winkler; Anna V. Kononova; Heike Trautmann; Tea Tušar; Penousal Machado; Thomas Bäck. 1. ed. Springer, Cham, 2024. p. 36-51 (Lecture Notes in Computer Science; Vol. 15151).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 12 - Chapter in an edited book (Author) › peer-review

TY - CHAP

T1 - LTR-HSS

T2 - A Learning-to-Rank Based Framework for Hypervolume Subset Selection

AU - Gong, Cheng

AU - Guo, Ping

AU - Shu, Tianye

AU - Zhang, Qingfu

AU - Ishibuchi, Hisao

PY - 2024/9/7

Y1 - 2024/9/7

N2 - Hypervolume subset selection (HSS) plays an important role in various aspects of the field of evolutionary multi-objective optimization, such as environmental selection and post-processing for decision-making. The goal of these problems is to find the optimal subset that maximizes the hypervolume from a given candidate solution set. Many methods have been developed to solve or approximately solve different types of HSS problems. However, existing approaches cannot effectively solve HSS problems with a large number of objectives within a short computation time. This drawback directly limits their applicability as a component for developing new EMO algorithms. In this paper, we propose a novel learning-to-rank based framework, named LTR-HSS, for solving the challenging HSS problems with a large number of objectives. The experimental results show that, compared to other state-of-the-art HSS methods, our proposed LTR-HSS requires a shorter computation time to solve HSS problems with large numbers of objectives while achieving superior or competitive hypervolume performance. This demonstrates the potential of our method to be integrated into algorithms for many-objective optimization. © 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG

AB - Hypervolume subset selection (HSS) plays an important role in various aspects of the field of evolutionary multi-objective optimization, such as environmental selection and post-processing for decision-making. The goal of these problems is to find the optimal subset that maximizes the hypervolume from a given candidate solution set. Many methods have been developed to solve or approximately solve different types of HSS problems. However, existing approaches cannot effectively solve HSS problems with a large number of objectives within a short computation time. This drawback directly limits their applicability as a component for developing new EMO algorithms. In this paper, we propose a novel learning-to-rank based framework, named LTR-HSS, for solving the challenging HSS problems with a large number of objectives. The experimental results show that, compared to other state-of-the-art HSS methods, our proposed LTR-HSS requires a shorter computation time to solve HSS problems with large numbers of objectives while achieving superior or competitive hypervolume performance. This demonstrates the potential of our method to be integrated into algorithms for many-objective optimization. © 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG

KW - Hypervolume subset selection

KW - Multi-objective optimization

KW - Machine learning

U2 - 10.1007/978-3-031-70085-9_3

DO - 10.1007/978-3-031-70085-9_3

M3 - RGC 12 - Chapter in an edited book (Author)

SN - 9783031700842

T3 - Lecture Notes in Computer Science

SP - 36

EP - 51

BT - Parallel Problem Solving from Nature

A2 - Affenzeller, Michael

A2 - Winkler, Stephan M.

A2 - Kononova, Anna V.

A2 - Trautmann, Heike

A2 - Tušar, Tea

A2 - Machado, Penousal

A2 - Bäck, Thomas

PB - Springer, Cham

ER -

Gong C , Guo P, Shu T, Zhang Q, Ishibuchi H. LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection. In Affenzeller M, Winkler SM, Kononova AV, Trautmann H, Tušar T, Machado P, Bäck T, editors, Parallel Problem Solving from Nature: PPSN XVII. 1 ed. Springer, Cham. 2024. p. 36-51. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-70085-9_3

LTR-HSS: A Learning-to-Rank Based Framework for Hypervolume Subset Selection

Abstract

Publication series

Funding

Research Keywords

RGC Funding Information

Access Output

Other Access Options

Permanent Link

Fingerprint

Projects

GRF: Few for Many: A Non-Pareto Approach for Many Objective Optimization

Cite this