Beyond Landscape Analysis: DynamoRep Features For Capturing Algorithm-Problem Interaction In Single-Objective Continuous Optimization

The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-insta...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Evolutionary computation s. 1
Hlavní autori: Cenikj, Gjorgjina, Petelin, Gašper, Doerr, Carola, Korošec, Peter, Eftimov, Tome
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States 07.03.2025
Predmet:
ISSN:1530-9304, 1530-9304
On-line prístup:Zistit podrobnosti o prístupe
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-instance algorithm selection and configuration approaches. Extending purely problem-centered feature collections, our recently proposed DynamoRep features provide a simple and inexpensive representation of the algorithmproblem interaction during the optimization process. In this paper, we conduct a comprehensive analysis of the predictive power of the DynamoRep features for the problem classification, algorithm selection, and algorithm classification tasks. In particular, the features are evaluated for the classification of problem instances into problem classes from the BBOB (Black Box Optimization Benchmarking) suite, selecting the best algorithm to solve a given problem from a portfolio of three algorithms (Differential Evolution, Evolutionary Strategy, and Particle Swarm Optimization), as well as distinguishing these algorithms based on their trajectories. We show that, despite being much cheaper to compute, they can yield results comparable to those using state-ofthe-art Exploratory Landscape Analysis features.
AbstractList The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-instance algorithm selection and configuration approaches. Extending purely problem-centered feature collections, our recently proposed DynamoRep features provide a simple and inexpensive representation of the algorithmproblem interaction during the optimization process. In this paper, we conduct a comprehensive analysis of the predictive power of the DynamoRep features for the problem classification, algorithm selection, and algorithm classification tasks. In particular, the features are evaluated for the classification of problem instances into problem classes from the BBOB (Black Box Optimization Benchmarking) suite, selecting the best algorithm to solve a given problem from a portfolio of three algorithms (Differential Evolution, Evolutionary Strategy, and Particle Swarm Optimization), as well as distinguishing these algorithms based on their trajectories. We show that, despite being much cheaper to compute, they can yield results comparable to those using state-ofthe-art Exploratory Landscape Analysis features.
The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-instance algorithm selection and configuration approaches. Extending purely problem-centered feature collections, our recently proposed DynamoRep features provide a simple and inexpensive representation of the algorithmproblem interaction during the optimization process. In this paper, we conduct a comprehensive analysis of the predictive power of the DynamoRep features for the problem classification, algorithm selection, and algorithm classification tasks. In particular, the features are evaluated for the classification of problem instances into problem classes from the BBOB (Black Box Optimization Benchmarking) suite, selecting the best algorithm to solve a given problem from a portfolio of three algorithms (Differential Evolution, Evolutionary Strategy, and Particle Swarm Optimization), as well as distinguishing these algorithms based on their trajectories. We show that, despite being much cheaper to compute, they can yield results comparable to those using state-ofthe-art Exploratory Landscape Analysis features.The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem instances, for analyzing the behavior of optimization algorithms, and the quality of existing problem benchmarks, as well as for automated per-instance algorithm selection and configuration approaches. Extending purely problem-centered feature collections, our recently proposed DynamoRep features provide a simple and inexpensive representation of the algorithmproblem interaction during the optimization process. In this paper, we conduct a comprehensive analysis of the predictive power of the DynamoRep features for the problem classification, algorithm selection, and algorithm classification tasks. In particular, the features are evaluated for the classification of problem instances into problem classes from the BBOB (Black Box Optimization Benchmarking) suite, selecting the best algorithm to solve a given problem from a portfolio of three algorithms (Differential Evolution, Evolutionary Strategy, and Particle Swarm Optimization), as well as distinguishing these algorithms based on their trajectories. We show that, despite being much cheaper to compute, they can yield results comparable to those using state-ofthe-art Exploratory Landscape Analysis features.
Author Petelin, Gašper
Cenikj, Gjorgjina
Eftimov, Tome
Doerr, Carola
Korošec, Peter
Author_xml – sequence: 1
  givenname: Gjorgjina
  surname: Cenikj
  fullname: Cenikj, Gjorgjina
  email: gjorgjina.cenikj@ijs.si
  organization: Jožef Stefan International Postgraduate School, Ljubljana, 1000, Slovenia gjorgjina.cenikj@ijs.si
– sequence: 2
  givenname: Gašper
  surname: Petelin
  fullname: Petelin, Gašper
  email: gasper.petelin@ijs.si
  organization: Jožef Stefan International Postgraduate School, Ljubljana, 1000, Slovenia gasper.petelin@ijs.si
– sequence: 3
  givenname: Carola
  surname: Doerr
  fullname: Doerr, Carola
  email: carola.doerr@lip6.fr
  organization: LIP6, Sorbonne Université, Paris, 75252, France carola.doerr@lip6.fr
– sequence: 4
  givenname: Peter
  surname: Korošec
  fullname: Korošec, Peter
  email: peter.korosec@ijs.si
  organization: Computer Systems Department, Jožef Stefan Institute, Ljubljana, 1000, Slovenia peter.korosec@ijs.si
– sequence: 5
  givenname: Tome
  surname: Eftimov
  fullname: Eftimov, Tome
  email: tome.eftimov@ijs.si
  organization: Computer Systems Department, Jožef Stefan Institute, Ljubljana, 1000, Slovenia tome.eftimov@ijs.si
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40053909$$D View this record in MEDLINE/PubMed
BookMark eNpNkE1LxDAQhoMo6qo3z5Kjl-okbfrhbV1dFRZW_DiXSXaqkTapTSusd_-3XVTwNM8MDy8vM2Hbzjti7FjAmRCpPKcP40ssAeIMtti-UDFERQzJ9j_eY5MQ3gBELEHssr0EQMUFFPvs65LW3q34At0qGGyJTx3W62DDBb9aO2z8A7V8TtgPHQU-9x2fYTsu1r3waf3iO9u_NtF953VNDb9zPXVoeuvdyPxxtGqKlvqNxtsH8Zl3vXWDHwJftr1t7Cdu3EO2U2Ed6Oh3HrDn-fXT7DZaLG_uZtNFZESqIJJoSJLWmZQYgypEpahQOaZ5sUIEXcUqNUrolIrc5Fle6So3SQaUaByxkAfs9Ce37fz7QKEvGxsM1TU6GjuVscgUZFIlG_XkVx10Q6uy7WyD3br8e538BlE9dVI
ContentType Journal Article
Copyright 2025 by the Massachusetts Institute of Technology.
Copyright_xml – notice: 2025 by the Massachusetts Institute of Technology.
DBID NPM
7X8
DOI 10.1162/evco_a_00370
DatabaseName PubMed
MEDLINE - Academic
DatabaseTitle PubMed
MEDLINE - Academic
DatabaseTitleList PubMed
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod no_fulltext_linktorsrc
Discipline Engineering
Computer Science
EISSN 1530-9304
ExternalDocumentID 40053909
Genre Journal Article
GroupedDBID ---
.DC
0R~
36B
4.4
53G
5GY
5VS
6IK
AAJGR
AAKMM
AALFJ
AALMD
AAYFX
ABAZT
ABDBF
ABJNI
ABVLG
ACM
ACUHS
ADL
AEBYY
AEJOY
AENEX
AENSD
AFWIH
AFWXC
AIKLT
AKRVB
ALMA_UNASSIGNED_HOLDINGS
ASPBG
AVWKF
AZFZN
BDXCO
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
EAP
EBS
EST
ESX
F5P
FEDTE
FNEHJ
GUFHI
HGAVV
HZ~
I07
IPLJI
JAVBF
LHSKQ
MCG
MINIK
NPM
O9-
OCL
P2P
PK0
RMI
ZWS
.4S
7X8
I-F
TUS
ID FETCH-LOGICAL-c1650-2ace2ebb722a30591f5e958a689daa0bf356c51b6e98c878fbf8c470e4babf892
IEDL.DBID 7X8
ISSN 1530-9304
IngestDate Fri Sep 05 13:02:31 EDT 2025
Mon Jul 21 05:59:28 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords algorithm classification
problem classification
single objective numerical optimization
feature construction
Language English
License 2025 by the Massachusetts Institute of Technology.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c1650-2ace2ebb722a30591f5e958a689daa0bf356c51b6e98c878fbf8c470e4babf892
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
OpenAccessLink https://hal.science/hal-05171409
PMID 40053909
PQID 3175072549
PQPubID 23479
ParticipantIDs proquest_miscellaneous_3175072549
pubmed_primary_40053909
PublicationCentury 2000
PublicationDate 2025-Mar-07
20250307
PublicationDateYYYYMMDD 2025-03-07
PublicationDate_xml – month: 03
  year: 2025
  text: 2025-Mar-07
  day: 07
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Evolutionary computation
PublicationTitleAlternate Evol Comput
PublicationYear 2025
SSID ssj0013201
Score 2.414825
Snippet The representation of optimization problems and algorithms in terms of numerical features is a well-established tool for comparing optimization problem...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 1
Title Beyond Landscape Analysis: DynamoRep Features For Capturing Algorithm-Problem Interaction In Single-Objective Continuous Optimization
URI https://www.ncbi.nlm.nih.gov/pubmed/40053909
https://www.proquest.com/docview/3175072549
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT9wwELZalkM5QIHyKBS5Uq8WTuLEMReEFlYcyu5KtNLeIj9hESQL-_gH_G9m8lC5VKrUS-RDLI3s8bxnPkJ-KGG5FjJhXhnNhHeKKRdFzIH74YS32hpTg03I4TCfTNS4DbjN27LKTibWgtpVFmPkp6jnuER35nz2zBA1CrOrLYTGR9JLwJTBki45eZ9F4O28VM4U-O1d4XsWn_qVrQqNkzEl_7txWSuZwdb_kveZbLbmJb1o-GGbfPDlDtnqoBto-5J3yMa7OYS75LXpY6E_se8XK6JoN6vkjF4iZH0FdjpFa3EJ3jkdVC-0r2eLusORXjzeASGL-yc2bsBpaB1lbBomYE1v4a9Hz0bmoRGuFCdiTctltZzTEUisp7YV9Av5Pbj61b9mLT4DsxEYdizW1sfeGBnHGsSGikLqVZrrLFdOa25CkmY2jUzmVW5zmQcTcisk98JoWKp4j6yVVekPCI1EcCYBJWCDE7hBJmnIguDO6SRT_pB87469AP7HpIYuPdBZ_Dn4Q7Lf3F0xawZ1FAJFjOLq6z_sPiKfYoT2xfIyeUx6AV6__0bW7Woxnb-c1IwF3-H45g1yjd3I
linkProvider ProQuest
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%3Ajournal&rft.genre=article&rft.atitle=Beyond+Landscape+Analysis%3A+DynamoRep+Features+For+Capturing+Algorithm-Problem+Interaction+In+Single-Objective+Continuous+Optimization&rft.jtitle=Evolutionary+computation&rft.au=Cenikj%2C+Gjorgjina&rft.au=Petelin%2C+Ga%C5%A1per&rft.au=Doerr%2C+Carola&rft.au=Koro%C5%A1ec%2C+Peter&rft.date=2025-03-07&rft.issn=1530-9304&rft.eissn=1530-9304&rft.spage=1&rft_id=info:doi/10.1162%2Fevco_a_00370&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1530-9304&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1530-9304&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1530-9304&client=summon