Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/36612
Title: Bi-objective Bayesian optimization of engineering problems with cheap and expensive cost functions
Authors: Loka, Nasrulloh
Couckuyt, Ivo
Garbuglia, Federico
Spina, Domenico
VAN NIEUWENHUYSE, Inneke 
Dhaene , Tom
Issue Date: 2023
Publisher: SPRINGER
Source: ENGINEERING WITH COMPUTERS, 9 , p. 1923 - 1933
Abstract: Multi-objective optimization of complex engineering systems is a challenging problem. The design goals can exhibit dynamic and nonlinear behaviour with respect to the system's parameters. Additionally, modern engineering is driven by simulation-based design which can be computationally expensive due to the complexity of the system under study. Bayesian optimization (BO) is a popular technique to tackle this kind of problem. In multi-objective BO, a data-driven surrogate model is created for each design objective. However, not all of the objectives may be expensive to compute. We develop an approach that can deal with a mix of expensive and cheap-to-evaluate objective functions. As a result, the proposed technique offers lower complexity than standard multi-objective BO methods and performs significantly better when the cheap objective function is difficult to approximate. In particular, we extend the popular hypervolume-based Expected Improvement (EI) and Probability of Improvement (POI) in bi-objective settings. The proposed methods are validated on multiple benchmark functions and two real-world engineering design optimization problems, showing that it performs better than its non-cheap counterparts. Furthermore, it performs competitively or better compared to other optimization methods.
Notes: Loka, N (corresponding author), Ghent Univ Imec, Dept Informat Technol INTEC, IDLab, iGent, Techno Pk Zwijnaarde 126, B-9052 Ghent, Belgium.
nasrulloh.loka@ugent.be
Keywords: Multi-objective optimization;Bayesian optimization;Hypervolume;Gaussian process
Document URI: http://hdl.handle.net/1942/36612
ISSN: 0177-0667
e-ISSN: 1435-5663
DOI: 10.1007/s00366-021-01573-7
ISI #: WOS:000746293500001
Rights: The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2022
Category: A1
Type: Journal Contribution
Validations: ecoom 2023
Appears in Collections:Research publications

Files in This Item:
File Description SizeFormat 
Bi-objective Bayesian optimization of engineering problems with cheap and expensive cost functions.pdf
  Restricted Access
Published version1.8 MBAdobe PDFView/Open    Request a copy
Cheap_Expensive_BO.pdfPeer-reviewed author version926 kBAdobe PDFView/Open
Show full item record

WEB OF SCIENCETM
Citations

6
checked on Mar 23, 2024

Page view(s)

68
checked on Sep 6, 2022

Download(s)

16
checked on Sep 6, 2022

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.