Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation

Jiangtao Shen; Peng Wang; Huachao Dong; Wenxin Wang

doi:10.1109/CEC53210.2023.10254133

Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation

Jiangtao Shen, Peng Wang, Huachao Dong, Wenxin Wang

School of Marine Science and Technology

Northwestern Polytechnical University Xian

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

1 Scopus citations

Abstract

Expensive many-objective optimization problems (EMaOPs) are common in the real world, whose objective values need to be calculated by time-consuming computational simulations or expensive experiments. For EMaOPs, guiding the optimization process by surrogate models is a popular method. In this paper, an FNN-assisted evolutionary algorithm is proposed for better solving EMaOPs. Concretely, a feedforward neural network (FNN) is constructed by learning the strengthened dominance relation (SDR) between two solutions. Then promising samples are generated by evolutionary search based on the constructed FNN. The proposed method is compared with four state-of-the-art peer algorithms on a set of benchmark problems. Experimental results demonstrate its superiority.

Original language	English
Title of host publication	2023 IEEE Congress on Evolutionary Computation, CEC 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350314588
DOIs	https://doi.org/10.1109/CEC53210.2023.10254133
State	Published - 2023
Event	2023 IEEE Congress on Evolutionary Computation, CEC 2023 - Chicago, United States Duration: 1 Jul 2023 → 5 Jul 2023

Publication series

Name	2023 IEEE Congress on Evolutionary Computation, CEC 2023

Conference

Conference	2023 IEEE Congress on Evolutionary Computation, CEC 2023
Country/Territory	United States
City	Chicago
Period	1/07/23 → 5/07/23

Keywords

evolutionary algorithm
expensive optimization
feedforward neural network
many-objective optimization

Access to Document

10.1109/CEC53210.2023.10254133

Cite this

@inproceedings{81a1d167b78c45169e926c4040af2ac9,

title = "Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation",

abstract = "Expensive many-objective optimization problems (EMaOPs) are common in the real world, whose objective values need to be calculated by time-consuming computational simulations or expensive experiments. For EMaOPs, guiding the optimization process by surrogate models is a popular method. In this paper, an FNN-assisted evolutionary algorithm is proposed for better solving EMaOPs. Concretely, a feedforward neural network (FNN) is constructed by learning the strengthened dominance relation (SDR) between two solutions. Then promising samples are generated by evolutionary search based on the constructed FNN. The proposed method is compared with four state-of-the-art peer algorithms on a set of benchmark problems. Experimental results demonstrate its superiority.",

keywords = "evolutionary algorithm, expensive optimization, feedforward neural network, many-objective optimization",

author = "Jiangtao Shen and Peng Wang and Huachao Dong and Wenxin Wang",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE Congress on Evolutionary Computation, CEC 2023 ; Conference date: 01-07-2023 Through 05-07-2023",

year = "2023",

doi = "10.1109/CEC53210.2023.10254133",

language = "英语",

series = "2023 IEEE Congress on Evolutionary Computation, CEC 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2023 IEEE Congress on Evolutionary Computation, CEC 2023",

}

Shen, J, Wang, P, Dong, H & Wang, W 2023, Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation. in 2023 IEEE Congress on Evolutionary Computation, CEC 2023. 2023 IEEE Congress on Evolutionary Computation, CEC 2023, Institute of Electrical and Electronics Engineers Inc., 2023 IEEE Congress on Evolutionary Computation, CEC 2023, Chicago, United States, 1/07/23. https://doi.org/10.1109/CEC53210.2023.10254133

Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation. / Shen, Jiangtao; Wang, Peng; Dong, Huachao et al.
2023 IEEE Congress on Evolutionary Computation, CEC 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (2023 IEEE Congress on Evolutionary Computation, CEC 2023).

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

TY - GEN

T1 - Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation

AU - Shen, Jiangtao

AU - Wang, Peng

AU - Dong, Huachao

AU - Wang, Wenxin

PY - 2023

Y1 - 2023

N2 - Expensive many-objective optimization problems (EMaOPs) are common in the real world, whose objective values need to be calculated by time-consuming computational simulations or expensive experiments. For EMaOPs, guiding the optimization process by surrogate models is a popular method. In this paper, an FNN-assisted evolutionary algorithm is proposed for better solving EMaOPs. Concretely, a feedforward neural network (FNN) is constructed by learning the strengthened dominance relation (SDR) between two solutions. Then promising samples are generated by evolutionary search based on the constructed FNN. The proposed method is compared with four state-of-the-art peer algorithms on a set of benchmark problems. Experimental results demonstrate its superiority.

AB - Expensive many-objective optimization problems (EMaOPs) are common in the real world, whose objective values need to be calculated by time-consuming computational simulations or expensive experiments. For EMaOPs, guiding the optimization process by surrogate models is a popular method. In this paper, an FNN-assisted evolutionary algorithm is proposed for better solving EMaOPs. Concretely, a feedforward neural network (FNN) is constructed by learning the strengthened dominance relation (SDR) between two solutions. Then promising samples are generated by evolutionary search based on the constructed FNN. The proposed method is compared with four state-of-the-art peer algorithms on a set of benchmark problems. Experimental results demonstrate its superiority.

KW - evolutionary algorithm

KW - expensive optimization

KW - feedforward neural network

KW - many-objective optimization

UR - http://www.scopus.com/inward/record.url?scp=85174532150&partnerID=8YFLogxK

U2 - 10.1109/CEC53210.2023.10254133

DO - 10.1109/CEC53210.2023.10254133

M3 - 会议稿件

AN - SCOPUS:85174532150

T3 - 2023 IEEE Congress on Evolutionary Computation, CEC 2023

BT - 2023 IEEE Congress on Evolutionary Computation, CEC 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE Congress on Evolutionary Computation, CEC 2023

Y2 - 1 July 2023 through 5 July 2023

ER -

Expensive Many-Objective Optimization by Learning of the Strengthened Dominance Relation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this