Towards Feature-Free Automated Algorithm Selection for Single-Objective Continuous Black-Box Optimization

We propose a novel method for automated algorithm selection in the domain of single-objective continuous black-box optimization. In contrast to existing methods, we use convolutional neural networks as the selection apparatus which bases its decision on a so-called 'fitness map'. This fitn...

Full description

Saved in:
Bibliographic Details
Published in:2021 IEEE Symposium Series on Computational Intelligence (SSCI) pp. 1 - 8
Main Authors: Prager, Raphael Patrick, Vinzent Seiler, Moritz, Trautmann, Heike, Kerschke, Pascal
Format: Conference Proceeding
Language:English
Published: IEEE 05.12.2021
Subjects:
Automated Algorithm Configuration
Benchmark testing
Black-Box Optimization
CMA-ES
Convolutional neural networks
Deep Learning
Feature-Free
Gray-scale
Performance gain
Point cloud compression
Reinforcement learning
Two dimensional displays
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract We propose a novel method for automated algorithm selection in the domain of single-objective continuous black-box optimization. In contrast to existing methods, we use convolutional neural networks as the selection apparatus which bases its decision on a so-called 'fitness map'. This fitness map is a 2D representation of a two dimensional search space where different gray scales indicate the quality of found solutions in certain areas. Our devised approach uses a modular CMA-ES framework which offers the option to create the conventional CMA-ES, CMA-ES with the alternate step-size adaptation and many other variants proposed over the years. In total, 4 608 different configurations are possible where most configurations are of complementary nature. In this proof-of-concept work, we consider a subset of 32 possible configurations. The developed method is evaluated against an excerpt of BBOB functions and its performance is compared against baselines that are commonly used in automated algorithm selection - the best standalone algorithm (configuration) and the best obtainable sequence of configurations. While the results indicate that the use of the fitness map is not superior on every benchmark problem, it indubitably shows its merit on more hard-to-solve problems. This offers a promising perspective for generalizing to other types of optimization problems and problem domains.
AbstractList We propose a novel method for automated algorithm selection in the domain of single-objective continuous black-box optimization. In contrast to existing methods, we use convolutional neural networks as the selection apparatus which bases its decision on a so-called 'fitness map'. This fitness map is a 2D representation of a two dimensional search space where different gray scales indicate the quality of found solutions in certain areas. Our devised approach uses a modular CMA-ES framework which offers the option to create the conventional CMA-ES, CMA-ES with the alternate step-size adaptation and many other variants proposed over the years. In total, 4 608 different configurations are possible where most configurations are of complementary nature. In this proof-of-concept work, we consider a subset of 32 possible configurations. The developed method is evaluated against an excerpt of BBOB functions and its performance is compared against baselines that are commonly used in automated algorithm selection - the best standalone algorithm (configuration) and the best obtainable sequence of configurations. While the results indicate that the use of the fitness map is not superior on every benchmark problem, it indubitably shows its merit on more hard-to-solve problems. This offers a promising perspective for generalizing to other types of optimization problems and problem domains.
Author Trautmann, Heike
Prager, Raphael Patrick
Vinzent Seiler, Moritz
Kerschke, Pascal
Author_xml – sequence: 1
  givenname: Raphael Patrick
  surname: Prager
  fullname: Prager, Raphael Patrick
  email: raphael.prager@wi.uni-muenster.de
  organization: Statistics and Optimization University of Münster,Münster,Germany
– sequence: 2
  givenname: Moritz
  surname: Vinzent Seiler
  fullname: Vinzent Seiler, Moritz
  email: moritz.seiler@wi.uni-muenster.de
  organization: Statistics and Optimization University of Münster,Münster,Germany
– sequence: 3
  givenname: Heike
  surname: Trautmann
  fullname: Trautmann, Heike
  email: trautmann@wi.uni-muenster.de
  organization: Statistics and Optimization University of Münster,Münster,Germany
– sequence: 4
  givenname: Pascal
  surname: Kerschke
  fullname: Kerschke, Pascal
  email: pascal.kerschke@tu-dresden.de
  organization: Big Data Analytics in Transportation TU Dresden,Dresden,Germany
BookMark eNotkE1ugzAUhF2pXbRpTlCp8gWgz8YYsySotJEisSBdRwYeqVvAkTH9O30SNauRZvSNNHNHrkc7IiGPDELGIH2qqnwdg4hZyIGzMJUSWCKuyDJNFEu4YikIpW6J2dpv7dqJFqj97DAoHCLNZm8H7bGlWb-3zvj3gVbYY-ONHWlnHa3MuO8xKOuPs_mFNLejN-Ns54muet18Biv7Q8uDN4P502fsntx0up9wedEFeSuet_lrsClf1nm2CQyHyAfIlRQt6gSaFholG4UxCtbJGiSoWEjRsVPSRbKpMY5V0gkO7WmQiECzGqMFefjvNYi4OzgzaPe7uxwQHQEDBFcT
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/SSCI50451.2021.9660174
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
EISBN 9781728190488
1728190487
EndPage 8
ExternalDocumentID 9660174
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i203t-e2864dea70cd0c86c8e5e41f6b06085464f1cd0f36cbe5587f420d281430a1be3
IEDL.DBID RIE
ISICitedReferencesCount 9
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000824464300350&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Thu Jun 29 18:37:51 EDT 2023
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-e2864dea70cd0c86c8e5e41f6b06085464f1cd0f36cbe5587f420d281430a1be3
PageCount 8
ParticipantIDs ieee_primary_9660174
PublicationCentury 2000
PublicationDate 2021-Dec.-5
PublicationDateYYYYMMDD 2021-12-05
PublicationDate_xml – month: 12
  year: 2021
  text: 2021-Dec.-5
  day: 05
PublicationDecade 2020
PublicationTitle 2021 IEEE Symposium Series on Computational Intelligence (SSCI)
PublicationTitleAbbrev SSCI
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.8989722
Snippet We propose a novel method for automated algorithm selection in the domain of single-objective continuous black-box optimization. In contrast to existing...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Automated Algorithm Configuration
Benchmark testing
Black-Box Optimization
CMA-ES
Convolutional neural networks
Deep Learning
Feature-Free
Gray-scale
Performance gain
Point cloud compression
Reinforcement learning
Two dimensional displays
Title Towards Feature-Free Automated Algorithm Selection for Single-Objective Continuous Black-Box Optimization
URI https://ieeexplore.ieee.org/document/9660174
WOSCitedRecordID wos000824464300350&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ3PS8MwFMfDNjx4UtnE3-Tg0Wxp1zbpcQ6HXrZBJ-w22uRFK3OVrhX_fF-yMhG8eCttSeElvHxe-r7vEXIL6OHiVCtmkIlYkPqCZQAeU7iYTKYilYJ2zSbEdCqXy3jeInd7LQwAuOQz6NtL9y9fF6q2R2UDW0kSCbpN2kKInVarEf16PB4kyfgptOVSMOrzvX7z8q-uKW7TmBz973PHpPejvqPz_b5yQlqw6ZJ84fJbt9QyW10Cm5QAdFRXBSInaDpavxQY57--08R1tkFzU-RRmuAQa2Cz7G3n2agtR5Vvagz4qTu8Y_fFF52h43hvFJk98jx5WIwfWdMmgeU-H1YMfBkFGlLBleZKRkpCCIFnooxHCFRBFBgPn5hhpDIIQylM4HPtSyQlnnoZDE9JZ1Ns4IxQpB3ppRLDQIkjeJDFXMdGhVwbIZENzknXmmn1sauEsWosdPH37UtyaGfCJX-EV6RTlTVckwP1WeXb8sZN3zcOGJ-M
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ3fS8MwEMfDnII-qWzib_Pgo9mSrj_SxzkcDuc22IS9jTa5amVrpWvFP99rViaCL76VtqRwCZfPpfe9I-QW0MP5gVYsQiZidmB5LAQQTOFiikLlqgC0aTbhjUZyPvcnNXK31cIAgEk-g1Z5af7l61QV5VFZu6wkiQS9Q3Yd27bERq1VyX4F99vTaW_glAVTMO6zRKt6_VffFLNt9A__98Ej0vzR39HJdmc5JjVIGiSemQzXNS2prciA9TMA2i3yFKETNO0uX1OM9N9WdGp626DBKRIpneIQS2Dj8H3j22hZkCpOCgz5qTm-Y_fpFx2j61hVmswmeek_zHqPrGqUwGKLd3IGlnRtDYHHleZKukqCA7aI3JC7iFS2a0cCn0QdV4XgONKLbItrSyIr8UCE0Dkh9SRN4JRQ5B0pAomBoMQRBIQ-136kHK4jTyIdnJFGaabFx6YWxqKy0Pnft2_I_uPsebgYDkZPF-SgnBWTCuJcknqeFXBF9tRnHq-zazOV34f1otM
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2021+IEEE+Symposium+Series+on+Computational+Intelligence+%28SSCI%29&rft.atitle=Towards+Feature-Free+Automated+Algorithm+Selection+for+Single-Objective+Continuous+Black-Box+Optimization&rft.au=Prager%2C+Raphael+Patrick&rft.au=Vinzent+Seiler%2C+Moritz&rft.au=Trautmann%2C+Heike&rft.au=Kerschke%2C+Pascal&rft.date=2021-12-05&rft.pub=IEEE&rft.spage=1&rft.epage=8&rft_id=info:doi/10.1109%2FSSCI50451.2021.9660174&rft.externalDocID=9660174