No Cover Image

Book chapter 1046 views 198 downloads

Towards Better Integration of Surrogate Models and Optimizers

Tinkle Chugh, Alma Rahat Orcid Logo, Vanessa Volz, Martin Zaefferer

High-Performance Simulation-Based Optimization, Volume: Chapter 7, Pages: 137 - 163

Swansea University Author: Alma Rahat Orcid Logo

Check full text

DOI (Published version): 10.1007/978-3-030-18764-4_7

Abstract

Surrogate-Assisted Evolutionary Algorithms (SAEAs) have been proven to be very effective in solving (synthetic and real-world) computationally expensive optimization problems with a limited number of function evaluations. The two main components of SAEAs are: the surrogate model and the evolutionary...

Full description

Published in: High-Performance Simulation-Based Optimization
ISBN: 9783030187637 9783030187644
ISSN: 1860-949X 1860-9503
Published: Cham Springer International Publishing 2020
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa52249
first_indexed 2019-10-02T20:21:34Z
last_indexed 2023-03-14T04:06:01Z
id cronfa52249
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2023-03-13T14:17:40.1222565</datestamp><bib-version>v2</bib-version><id>52249</id><entry>2019-10-02</entry><title>Towards Better Integration of Surrogate Models and Optimizers</title><swanseaauthors><author><sid>6206f027aca1e3a5ff6b8cd224248bc2</sid><ORCID>0000-0002-5023-1371</ORCID><firstname>Alma</firstname><surname>Rahat</surname><name>Alma Rahat</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2019-10-02</date><deptcode>MACS</deptcode><abstract>Surrogate-Assisted Evolutionary Algorithms (SAEAs) have been proven to be very effective in solving (synthetic and real-world) computationally expensive optimization problems with a limited number of function evaluations. The two main components of SAEAs are: the surrogate model and the evolutionary optimizer, both of which use parameters to control their respective behavior. These parameters are likely to interact closely, and hence the exploitation of any such relationships may lead to the design of an enhanced SAEA. In this chapter, as a first step, we focus on Kriging and the Efficient Global Optimization (EGO) framework. We discuss potentially profitable ways of a better integration of model and optimizer. Furthermore, we investigate in depth how different parameters of the model and the optimizer impact optimization results. In particular, we determine whether there are any interactions between these parameters, and how the problem characteristics impact optimization results. In the experimental study, we use the popular Black-Box Optimization Benchmarking (BBOB) testbed. Interestingly, the analysis finds no evidence for significant interactions between model and optimizer parameters, but independently their performance has a significant interaction with the objective function. Based on our results, we make recommendations on how best to configure EGO.</abstract><type>Book chapter</type><journal>High-Performance Simulation-Based Optimization</journal><volume>Chapter 7</volume><journalNumber/><paginationStart>137</paginationStart><paginationEnd>163</paginationEnd><publisher>Springer International Publishing</publisher><placeOfPublication>Cham</placeOfPublication><isbnPrint>9783030187637</isbnPrint><isbnElectronic>9783030187644</isbnElectronic><issnPrint>1860-949X</issnPrint><issnElectronic>1860-9503</issnElectronic><keywords/><publishedDay>1</publishedDay><publishedMonth>1</publishedMonth><publishedYear>2020</publishedYear><publishedDate>2020-01-01</publishedDate><doi>10.1007/978-3-030-18764-4_7</doi><url>http://dx.doi.org/10.1007/978-3-030-18764-4_7</url><notes/><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm/><funders/><projectreference/><lastEdited>2023-03-13T14:17:40.1222565</lastEdited><Created>2019-10-02T15:16:46.7168270</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Tinkle</firstname><surname>Chugh</surname><order>1</order></author><author><firstname>Alma</firstname><surname>Rahat</surname><orcid>0000-0002-5023-1371</orcid><order>2</order></author><author><firstname>Vanessa</firstname><surname>Volz</surname><order>3</order></author><author><firstname>Martin</firstname><surname>Zaefferer</surname><order>4</order></author></authors><documents><document><filename>52249__15868__2f8ddc1377334accaf3d639b162495df.pdf</filename><originalFilename>52249.pdf</originalFilename><uploaded>2019-11-12T16:21:34.1153215</uploaded><type>Output</type><contentLength>316180</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2020-06-02T00:00:00.0000000</embargoDate><copyrightCorrect>false</copyrightCorrect></document></documents><OutputDurs/></rfc1807>
spelling 2023-03-13T14:17:40.1222565 v2 52249 2019-10-02 Towards Better Integration of Surrogate Models and Optimizers 6206f027aca1e3a5ff6b8cd224248bc2 0000-0002-5023-1371 Alma Rahat Alma Rahat true false 2019-10-02 MACS Surrogate-Assisted Evolutionary Algorithms (SAEAs) have been proven to be very effective in solving (synthetic and real-world) computationally expensive optimization problems with a limited number of function evaluations. The two main components of SAEAs are: the surrogate model and the evolutionary optimizer, both of which use parameters to control their respective behavior. These parameters are likely to interact closely, and hence the exploitation of any such relationships may lead to the design of an enhanced SAEA. In this chapter, as a first step, we focus on Kriging and the Efficient Global Optimization (EGO) framework. We discuss potentially profitable ways of a better integration of model and optimizer. Furthermore, we investigate in depth how different parameters of the model and the optimizer impact optimization results. In particular, we determine whether there are any interactions between these parameters, and how the problem characteristics impact optimization results. In the experimental study, we use the popular Black-Box Optimization Benchmarking (BBOB) testbed. Interestingly, the analysis finds no evidence for significant interactions between model and optimizer parameters, but independently their performance has a significant interaction with the objective function. Based on our results, we make recommendations on how best to configure EGO. Book chapter High-Performance Simulation-Based Optimization Chapter 7 137 163 Springer International Publishing Cham 9783030187637 9783030187644 1860-949X 1860-9503 1 1 2020 2020-01-01 10.1007/978-3-030-18764-4_7 http://dx.doi.org/10.1007/978-3-030-18764-4_7 COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University 2023-03-13T14:17:40.1222565 2019-10-02T15:16:46.7168270 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Tinkle Chugh 1 Alma Rahat 0000-0002-5023-1371 2 Vanessa Volz 3 Martin Zaefferer 4 52249__15868__2f8ddc1377334accaf3d639b162495df.pdf 52249.pdf 2019-11-12T16:21:34.1153215 Output 316180 application/pdf Accepted Manuscript true 2020-06-02T00:00:00.0000000 false
title Towards Better Integration of Surrogate Models and Optimizers
spellingShingle Towards Better Integration of Surrogate Models and Optimizers
Alma Rahat
title_short Towards Better Integration of Surrogate Models and Optimizers
title_full Towards Better Integration of Surrogate Models and Optimizers
title_fullStr Towards Better Integration of Surrogate Models and Optimizers
title_full_unstemmed Towards Better Integration of Surrogate Models and Optimizers
title_sort Towards Better Integration of Surrogate Models and Optimizers
author_id_str_mv 6206f027aca1e3a5ff6b8cd224248bc2
author_id_fullname_str_mv 6206f027aca1e3a5ff6b8cd224248bc2_***_Alma Rahat
author Alma Rahat
author2 Tinkle Chugh
Alma Rahat
Vanessa Volz
Martin Zaefferer
format Book chapter
container_title High-Performance Simulation-Based Optimization
container_volume Chapter 7
container_start_page 137
publishDate 2020
institution Swansea University
isbn 9783030187637
9783030187644
issn 1860-949X
1860-9503
doi_str_mv 10.1007/978-3-030-18764-4_7
publisher Springer International Publishing
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
url http://dx.doi.org/10.1007/978-3-030-18764-4_7
document_store_str 1
active_str 0
description Surrogate-Assisted Evolutionary Algorithms (SAEAs) have been proven to be very effective in solving (synthetic and real-world) computationally expensive optimization problems with a limited number of function evaluations. The two main components of SAEAs are: the surrogate model and the evolutionary optimizer, both of which use parameters to control their respective behavior. These parameters are likely to interact closely, and hence the exploitation of any such relationships may lead to the design of an enhanced SAEA. In this chapter, as a first step, we focus on Kriging and the Efficient Global Optimization (EGO) framework. We discuss potentially profitable ways of a better integration of model and optimizer. Furthermore, we investigate in depth how different parameters of the model and the optimizer impact optimization results. In particular, we determine whether there are any interactions between these parameters, and how the problem characteristics impact optimization results. In the experimental study, we use the popular Black-Box Optimization Benchmarking (BBOB) testbed. Interestingly, the analysis finds no evidence for significant interactions between model and optimizer parameters, but independently their performance has a significant interaction with the objective function. Based on our results, we make recommendations on how best to configure EGO.
published_date 2020-01-01T05:01:08Z
_version_ 1822105134727954432
score 11.2862625

Similar Items