A new acquisition function for robust Bayesian optimization of unconstrained problems

Sibghat Ullah; Hao Wang; Stefan Menzel; Bernhard Sendhoff; Thomas Bäck

doi:10.1145/3449726.3463206

A new acquisition function for robust Bayesian optimization of unconstrained problems

Sibghat Ullah, Hao Wang, Stefan Menzel, Bernhard Sendhoff, Thomas Bäck

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

49 Downloads (Pure)

Abstract

A new acquisition function is proposed for solving robust optimization problems via Bayesian Optimization. The proposed acquisition function reflects the need for the robust instead of the nominal optimum, and is based on the intuition of utilizing the higher moments of the improvement. The efficacy of Bayesian Optimization based on this acquisition function is demonstrated on four test problems, each affected by three different levels of noise. Our findings suggest the promising nature of the proposed acquisition function as it yields a better robust optimal value of the function in 6/12 test scenarios when compared with the baseline.

Original language	English
Title of host publication	GECCO '21
Subtitle of host publication	Proceedings of the Genetic and Evolutionary Computation Conference Companion
Editors	Francisco Chicano
Place of Publication	New York
Publisher	Association for Computing Machinery (ACM)
Pages	1344-1345
Number of pages	2
ISBN (Electronic)	9781450383516
DOIs	https://doi.org/10.1145/3449726.3463206
Publication status	Published - 7 Jul 2021
Externally published	Yes
Event	2021 Genetic and Evolutionary Computation Conference, GECCO 2021 - Virtual, Online, France Duration: 10 Jul 2021 → 14 Jul 2021

Publication series

Name	Genetic and Evolutionary Computation Conference (GECCO)
Publisher	Association for Computing Machinery (ACM)

Conference

Conference	2021 Genetic and Evolutionary Computation Conference, GECCO 2021
Country/Territory	France
City	Virtual, Online
Period	10/07/21 → 14/07/21

Bibliographical note

Funding Information:
This research has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement number 766186 (ECOLE).

Publisher Copyright:
© 2021 ACM.

Keywords

Bayesian optimization
kriging
robust optimization

ASJC Scopus subject areas

Computer Science Applications
Software
Computational Theory and Mathematics

Access to Document

10.1145/3449726.3463206Licence: None: All rights reserved

UllahS2021AnewacFinal published version, 1.19 MBLicence: None: All rights reserved

H2020_ITN_ECOLE_Coordinator
Yao, X. (Principal Investigator)
European Commission, European Commission - Management Costs
1/04/18 → 31/03/22
Project: Research

Cite this

Ullah, S., Wang, H., Menzel, S., Sendhoff, B., & Bäck, T. (2021). A new acquisition function for robust Bayesian optimization of unconstrained problems. In F. Chicano (Ed.), GECCO '21: Proceedings of the Genetic and Evolutionary Computation Conference Companion (pp. 1344-1345). (Genetic and Evolutionary Computation Conference (GECCO)). Association for Computing Machinery (ACM). https://doi.org/10.1145/3449726.3463206

Ullah, Sibghat ; Wang, Hao ; Menzel, Stefan et al. / A new acquisition function for robust Bayesian optimization of unconstrained problems. GECCO '21: Proceedings of the Genetic and Evolutionary Computation Conference Companion. editor / Francisco Chicano. New York : Association for Computing Machinery (ACM), 2021. pp. 1344-1345 (Genetic and Evolutionary Computation Conference (GECCO)).

@inproceedings{857ec6e60aed45d6924c227e2cb8f7fa,

title = "A new acquisition function for robust Bayesian optimization of unconstrained problems",

abstract = "A new acquisition function is proposed for solving robust optimization problems via Bayesian Optimization. The proposed acquisition function reflects the need for the robust instead of the nominal optimum, and is based on the intuition of utilizing the higher moments of the improvement. The efficacy of Bayesian Optimization based on this acquisition function is demonstrated on four test problems, each affected by three different levels of noise. Our findings suggest the promising nature of the proposed acquisition function as it yields a better robust optimal value of the function in 6/12 test scenarios when compared with the baseline. ",

keywords = "Bayesian optimization, kriging, robust optimization",

author = "Sibghat Ullah and Hao Wang and Stefan Menzel and Bernhard Sendhoff and Thomas B{\"a}ck",

note = "Funding Information: This research has received funding from the European Union{\textquoteright}s Horizon 2020 research and innovation programme under grant agreement number 766186 (ECOLE). Publisher Copyright: {\textcopyright} 2021 ACM.; 2021 Genetic and Evolutionary Computation Conference, GECCO 2021 ; Conference date: 10-07-2021 Through 14-07-2021",

year = "2021",

month = jul,

day = "7",

doi = "10.1145/3449726.3463206",

language = "English",

series = "Genetic and Evolutionary Computation Conference (GECCO)",

publisher = "Association for Computing Machinery (ACM)",

pages = "1344--1345",

editor = "Francisco Chicano",

booktitle = "GECCO '21",

address = "United States",

}

Ullah, S, Wang, H, Menzel, S, Sendhoff, B & Bäck, T 2021, A new acquisition function for robust Bayesian optimization of unconstrained problems. in F Chicano (ed.), GECCO '21: Proceedings of the Genetic and Evolutionary Computation Conference Companion. Genetic and Evolutionary Computation Conference (GECCO), Association for Computing Machinery (ACM), New York, pp. 1344-1345, 2021 Genetic and Evolutionary Computation Conference, GECCO 2021, Virtual, Online, France, 10/07/21. https://doi.org/10.1145/3449726.3463206

A new acquisition function for robust Bayesian optimization of unconstrained problems. / Ullah, Sibghat; Wang, Hao; Menzel, Stefan et al.
GECCO '21: Proceedings of the Genetic and Evolutionary Computation Conference Companion. ed. / Francisco Chicano. New York: Association for Computing Machinery (ACM), 2021. p. 1344-1345 (Genetic and Evolutionary Computation Conference (GECCO)).

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

TY - GEN

T1 - A new acquisition function for robust Bayesian optimization of unconstrained problems

AU - Ullah, Sibghat

AU - Wang, Hao

AU - Menzel, Stefan

AU - Sendhoff, Bernhard

AU - Bäck, Thomas

N1 - Funding Information: This research has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement number 766186 (ECOLE). Publisher Copyright: © 2021 ACM.

PY - 2021/7/7

Y1 - 2021/7/7

N2 - A new acquisition function is proposed for solving robust optimization problems via Bayesian Optimization. The proposed acquisition function reflects the need for the robust instead of the nominal optimum, and is based on the intuition of utilizing the higher moments of the improvement. The efficacy of Bayesian Optimization based on this acquisition function is demonstrated on four test problems, each affected by three different levels of noise. Our findings suggest the promising nature of the proposed acquisition function as it yields a better robust optimal value of the function in 6/12 test scenarios when compared with the baseline.

AB - A new acquisition function is proposed for solving robust optimization problems via Bayesian Optimization. The proposed acquisition function reflects the need for the robust instead of the nominal optimum, and is based on the intuition of utilizing the higher moments of the improvement. The efficacy of Bayesian Optimization based on this acquisition function is demonstrated on four test problems, each affected by three different levels of noise. Our findings suggest the promising nature of the proposed acquisition function as it yields a better robust optimal value of the function in 6/12 test scenarios when compared with the baseline.

KW - Bayesian optimization

KW - kriging

KW - robust optimization

UR - http://www.scopus.com/inward/record.url?scp=85111015437&partnerID=8YFLogxK

U2 - 10.1145/3449726.3463206

DO - 10.1145/3449726.3463206

M3 - Conference contribution

AN - SCOPUS:85111015437

T3 - Genetic and Evolutionary Computation Conference (GECCO)

SP - 1344

EP - 1345

BT - GECCO '21

A2 - Chicano, Francisco

PB - Association for Computing Machinery (ACM)

CY - New York

T2 - 2021 Genetic and Evolutionary Computation Conference, GECCO 2021

Y2 - 10 July 2021 through 14 July 2021

ER -

Ullah S, Wang H, Menzel S, Sendhoff B, Bäck T. A new acquisition function for robust Bayesian optimization of unconstrained problems. In Chicano F, editor, GECCO '21: Proceedings of the Genetic and Evolutionary Computation Conference Companion. New York: Association for Computing Machinery (ACM). 2021. p. 1344-1345. (Genetic and Evolutionary Computation Conference (GECCO)). doi: 10.1145/3449726.3463206

A new acquisition function for robust Bayesian optimization of unconstrained problems

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Projects

H2020_ITN_ECOLE_Coordinator

Cite this