Data Mining Approach for Feature Based Parameter Tunning for Mixed-Integer Programming Solvers

Integer Programming (IP) is the most successful technique for solving hard combinatorial optimization problems. Modern IP solvers are very complex programs composed of many different procedures whose execution is embedded in the generic Branch & Bound framework. The activation of these procedure...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Procedia computer science Ročník 108; s. 715 - 724
Hlavní autoři: Boas, Matheus G. Vilas, Santos, Haroldo G., Martins, Rafael de S.O., Merschmann, Luiz H.C.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 2017
Témata:
ISSN:1877-0509, 1877-0509
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Integer Programming (IP) is the most successful technique for solving hard combinatorial optimization problems. Modern IP solvers are very complex programs composed of many different procedures whose execution is embedded in the generic Branch & Bound framework. The activation of these procedures as well the definition of exploration strategies for the search tree can be done by setting different parameters. Since the success of these procedures and strategies in improving the performance of IP solvers varies widely depending on the problem being solved, the usual approach for discovering a good set of parameters considering average results is not ideal. In this work we propose a comprehensive approach for the automatic tuning of Integer Programming solvers where the characteristics of instances are considered. Computational experiments in a diverse set of 308 benchmark instances using the open source COIN-OR CBC solver were performed with different parameter sets and the results were processed by data mining algorithms. The results were encouraging: when trained with a portion of the database the algorithms were able to predict better parameters for the remaining instances in 84% of the cases. The selection of a single best parameter setting would provide an improvement in only 56% of instances, showing that great improvements can be obtained with our approach.
AbstractList Integer Programming (IP) is the most successful technique for solving hard combinatorial optimization problems. Modern IP solvers are very complex programs composed of many different procedures whose execution is embedded in the generic Branch & Bound framework. The activation of these procedures as well the definition of exploration strategies for the search tree can be done by setting different parameters. Since the success of these procedures and strategies in improving the performance of IP solvers varies widely depending on the problem being solved, the usual approach for discovering a good set of parameters considering average results is not ideal. In this work we propose a comprehensive approach for the automatic tuning of Integer Programming solvers where the characteristics of instances are considered. Computational experiments in a diverse set of 308 benchmark instances using the open source COIN-OR CBC solver were performed with different parameter sets and the results were processed by data mining algorithms. The results were encouraging: when trained with a portion of the database the algorithms were able to predict better parameters for the remaining instances in 84% of the cases. The selection of a single best parameter setting would provide an improvement in only 56% of instances, showing that great improvements can be obtained with our approach.
Author Boas, Matheus G. Vilas
Santos, Haroldo G.
Merschmann, Luiz H.C.
Martins, Rafael de S.O.
Author_xml – sequence: 1
  givenname: Matheus G. Vilas
  surname: Boas
  fullname: Boas, Matheus G. Vilas
  email: matheusgueedes91@gmail.com
  organization: Federal University of Ouro Preto, Department of Computing, Ouro Preto, Brazil
– sequence: 2
  givenname: Haroldo G.
  surname: Santos
  fullname: Santos, Haroldo G.
  email: haroldo.santos@gmail.com
  organization: Federal University of Ouro Preto, Department of Computing, Ouro Preto, Brazil
– sequence: 3
  givenname: Rafael de S.O.
  surname: Martins
  fullname: Martins, Rafael de S.O.
  email: martins.rso@gmail.com
  organization: Federal University of Ouro Preto, Department of Information Systems and Computing, João Monlevade, Brazil
– sequence: 4
  givenname: Luiz H.C.
  surname: Merschmann
  fullname: Merschmann, Luiz H.C.
  email: luiz.hcm@dcc.ufla.br
  organization: Federal University of Lavras, Department of Computer Science, Lavras, Brazil
BookMark eNqFkEFPAjEQhRuDiYj8Ai_7B3Ztd-m2PXhAFCXBSCJebcZ2Fktgl7SF6L93AQ_Gg85lJnnve8m8c9KpmxoJuWQ0Y5SVV8ts4xsTspwykVGe5bI8IV0mhUgpp6rz4z4j_RCWtJ1CSsVEl7zeQoTk0dWuXiTDTZsE5j2pGp-MEeLWY3IDAW0yAw9rjOiT-bY-mPeeR_eBNp3UERetMvPNonWt9-pzs9qhDxfktIJVwP737pGX8d189JBOn-4no-E0NYUqY8qkFcxU1ihlRMUZDJAXYLlCBcYCrzA3WDJUVua8EAiQywGWXOTyzeTGFD1SHHONb0LwWOmNd2vwn5pRvW9JL_WhJb1vSVOu25ZaSv2ijIsQXVNHD271D3t9ZLF9a-fQ62Ac1gat82iito37k_8CWgCIug
CitedBy_id crossref_primary_10_1016_j_endm_2018_03_029
Cites_doi 10.1016/j.ejor.2013.10.043
10.1147/rd.471.0057
10.1371/journal.pcbi.0020094
10.1145/1068009.1068194
10.2307/1910129
10.1007/s12532-011-0025-9
10.1109/72.870050
10.1137/040620886
10.1016/j.cor.2015.07.002
10.1023/A:1006559212014
ContentType Journal Article
Copyright 2017
Copyright_xml – notice: 2017
DBID 6I.
AAFTH
AAYXX
CITATION
DOI 10.1016/j.procs.2017.05.286
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 1877-0509
EndPage 724
ExternalDocumentID 10_1016_j_procs_2017_05_286
S1877050917309389
GroupedDBID --K
0R~
0SF
1B1
457
5VS
6I.
71M
AACTN
AAEDT
AAEDW
AAFTH
AAIKJ
AALRI
AAQFI
AAXUO
ABMAC
ACGFS
ADBBV
ADEZE
AEXQZ
AFTJW
AGHFR
AITUG
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
E3Z
EBS
EJD
EP3
FDB
FNPLU
HZ~
IXB
KQ8
M41
M~E
NCXOZ
O-L
O9-
OK1
P2P
RIG
ROL
SES
SSZ
9DU
AAYWO
AAYXX
ABWVN
ACRPL
ACVFH
ADCNI
ADNMO
ADVLN
AEUPX
AFPUW
AIGII
AKBMS
AKRWK
AKYEP
CITATION
~HD
ID FETCH-LOGICAL-c396t-18d71cfdc99c7f51a4e53ad59e9acda5fe2ce61e9d82537eaa284e65728bc2cc3
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000404959000072&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1877-0509
IngestDate Sat Nov 29 04:09:41 EST 2025
Tue Nov 18 21:17:20 EST 2025
Wed May 17 00:58:04 EDT 2023
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords mixed-integer programming
cut
coin-or branch
regression algorithms
data mining
feature based parameter tunning
Language English
License This is an open access article under the CC BY-NC-ND license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c396t-18d71cfdc99c7f51a4e53ad59e9acda5fe2ce61e9d82537eaa284e65728bc2cc3
OpenAccessLink https://dx.doi.org/10.1016/j.procs.2017.05.286
PageCount 10
ParticipantIDs crossref_primary_10_1016_j_procs_2017_05_286
crossref_citationtrail_10_1016_j_procs_2017_05_286
elsevier_sciencedirect_doi_10_1016_j_procs_2017_05_286
PublicationCentury 2000
PublicationDate 2017
2017-00-00
PublicationDateYYYYMMDD 2017-01-01
PublicationDate_xml – year: 2017
  text: 2017
PublicationDecade 2010
PublicationTitle Procedia computer science
PublicationYear 2017
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Lougee-Heimer (bib0013) 2003; 47
Joseph Haas, Maxim Peysakhov, and Spiros Mancoridis. Ga-based Parameter Tuning for Multi-agent Systems. In
Vapnik (bib0016) 1995
Kohavi, John (bib0011) 1995
Lpez-Ibez, Sttzle (bib0014) 2014; 235
Bellio, Ceschia, Gaspero, Schaerf, Urli (bib0005) 2016; 65
Achard, De Schutter (bib0001) 07 2006; 2
Mustafa Baz, Brady Hunsaker, P Brooks, and Abhijit Gosavi. Automated Tuning of Optimization Software Parameters. Technical report, Technical Report TR2007-7, University of Pittsburgh, Department of Industrial Engineering, 2007.
Land, Doig (bib0012) 1960; 28
Bixby, Fenelon, Gu, Rothberg, Wunderling (bib0006) 2004
Koch, Achterberg, Andersen, Bastert, Berthold, Bixby, Danna, Gamrath, Gleixner, Heinz (bib0010) 2011; 3
Michael R. Garey and David S. Johnson.
GECCO ‘05, pages 1085–1086, New York, NY, USA, 2005. ACM.
Witten, Frank, Hall (bib0017) 2011
Shevade, Keerthi, Bhattacharyya, Murthy (bib0015) 2000; 11
Atkeson, Moore, Schaal (bib0002) 1997
Forrest, Lougee-Heimer (bib0007) 2005
W. H. Freeman & Co., New York, NY, USA, 1979.
Audet, Orban (bib0003) 2006; 17
Bixby (10.1016/j.procs.2017.05.286_bib0006) 2004
Forrest (10.1016/j.procs.2017.05.286_bib0007) 2005
10.1016/j.procs.2017.05.286_bib0004
Witten (10.1016/j.procs.2017.05.286_bib0017) 2011
Achard (10.1016/j.procs.2017.05.286_bib0001) 2006; 2
10.1016/j.procs.2017.05.286_bib0009
10.1016/j.procs.2017.05.286_bib0008
Land (10.1016/j.procs.2017.05.286_bib0012) 1960; 28
Lougee-Heimer (10.1016/j.procs.2017.05.286_bib0013) 2003; 47
Lpez-Ibez (10.1016/j.procs.2017.05.286_bib0014) 2014; 235
Audet (10.1016/j.procs.2017.05.286_bib0003) 2006; 17
Bellio (10.1016/j.procs.2017.05.286_bib0005) 2016; 65
Koch (10.1016/j.procs.2017.05.286_bib0010) 2011; 3
Shevade (10.1016/j.procs.2017.05.286_bib0015) 2000; 11
Kohavi (10.1016/j.procs.2017.05.286_bib0011) 1995
Atkeson (10.1016/j.procs.2017.05.286_bib0002) 1997
Vapnik (10.1016/j.procs.2017.05.286_bib0016) 1995
References_xml – volume: 17
  start-page: 642
  year: 2006
  end-page: 664
  ident: bib0003
  article-title: Finding Optimal Algorithmic Parameters Using Derivative-Free Optimization
  publication-title: SIAM Journal on Optimization
– year: 1995
  ident: bib0016
  publication-title: The Nature of Statistical Learning Theory
– volume: 235
  start-page: 569
  year: 2014
  end-page: 582
  ident: bib0014
  article-title: Automatically Improving the Anytime Behaviour of Optimisation Algorithms
  publication-title: European Journal of Operational Research
– year: 2011
  ident: bib0017
  publication-title: In Ian H. Witten, Eibe Frank, and Mark A. Hall, editors, Data Mining: Practical Machine Learning Tools and Techniques
– volume: 65
  start-page: 83
  year: 2016
  end-page: 92
  ident: bib0005
  article-title: Feature-based Tuning of Simulated Annealing applied to the Curriculum-based Course Timetabling Problem
  publication-title: Computers & Operations Research
– reference: W. H. Freeman & Co., New York, NY, USA, 1979.
– reference: Michael R. Garey and David S. Johnson.
– volume: 3
  year: 2011
  ident: bib0010
  article-title: MIPLIB 2010
  publication-title: Mathematical Programming Computation
– volume: 47
  start-page: 57
  year: 2003
  end-page: 66
  ident: bib0013
  article-title: The Common Optimization Interface for Operations Research: Promoting Open-source Software in the Operations Research Community
  publication-title: IBM Journal of Research and Development
– start-page: 309
  year: 2004
  end-page: 325
  ident: bib0006
  article-title: Mixed Integer Programming: A Progress Report
  publication-title: The sharpest cut: the impact of Manfred Padberg and his work, chapter 18
– reference: , GECCO ‘05, pages 1085–1086, New York, NY, USA, 2005. ACM.
– volume: 2
  start-page: 1
  year: 07 2006
  end-page: 11
  ident: bib0001
  article-title: Complex Parameter Landscape for a Complex Neuron Model
  publication-title: PLOS Computational Biology
– reference: Mustafa Baz, Brady Hunsaker, P Brooks, and Abhijit Gosavi. Automated Tuning of Optimization Software Parameters. Technical report, Technical Report TR2007-7, University of Pittsburgh, Department of Industrial Engineering, 2007.
– start-page: 257
  year: 2005
  end-page: 277
  ident: bib0007
  article-title: Cbc User Guide
  publication-title: Emerging Theory, Methods, and Applications
– reference: Joseph Haas, Maxim Peysakhov, and Spiros Mancoridis. Ga-based Parameter Tuning for Multi-agent Systems. In
– volume: 11
  year: 2000
  ident: bib0015
  article-title: Improvements to the SMO Algorithm for SVM Regression
  publication-title: IEEE Transactions on Neural Networks
– start-page: 11
  year: 1997
  end-page: 73
  ident: bib0002
  article-title: Locally Weighted Learning
  publication-title: Artificial Intelligence Review
– start-page: 304
  year: 1995
  end-page: 312
  ident: bib0011
  article-title: Automatic Parameter Selection by Minimizing Estimated Error
  publication-title: In Proceedings of the Twelfth International Conference on Machine Learning
– volume: 28
  start-page: 497
  year: 1960
  end-page: 520
  ident: bib0012
  article-title: An Automatic Method for Solving Discrete Programming Problems
  publication-title: Econometrica
– ident: 10.1016/j.procs.2017.05.286_bib0004
– volume: 235
  start-page: 569
  issue: 3
  year: 2014
  ident: 10.1016/j.procs.2017.05.286_bib0014
  article-title: Automatically Improving the Anytime Behaviour of Optimisation Algorithms
  publication-title: European Journal of Operational Research
  doi: 10.1016/j.ejor.2013.10.043
– volume: 47
  start-page: 57
  issue: 1
  year: 2003
  ident: 10.1016/j.procs.2017.05.286_bib0013
  article-title: The Common Optimization Interface for Operations Research: Promoting Open-source Software in the Operations Research Community
  publication-title: IBM Journal of Research and Development
  doi: 10.1147/rd.471.0057
– volume: 2
  start-page: 1
  issue: 7
  year: 2006
  ident: 10.1016/j.procs.2017.05.286_bib0001
  article-title: Complex Parameter Landscape for a Complex Neuron Model
  publication-title: PLOS Computational Biology
  doi: 10.1371/journal.pcbi.0020094
– year: 1995
  ident: 10.1016/j.procs.2017.05.286_bib0016
– ident: 10.1016/j.procs.2017.05.286_bib0008
– ident: 10.1016/j.procs.2017.05.286_bib0009
  doi: 10.1145/1068009.1068194
– volume: 28
  start-page: 497
  year: 1960
  ident: 10.1016/j.procs.2017.05.286_bib0012
  article-title: An Automatic Method for Solving Discrete Programming Problems
  publication-title: Econometrica
  doi: 10.2307/1910129
– volume: 3
  issue: 2
  year: 2011
  ident: 10.1016/j.procs.2017.05.286_bib0010
  article-title: MIPLIB 2010
  publication-title: Mathematical Programming Computation
  doi: 10.1007/s12532-011-0025-9
– start-page: 304
  year: 1995
  ident: 10.1016/j.procs.2017.05.286_bib0011
  article-title: Automatic Parameter Selection by Minimizing Estimated Error
– year: 2011
  ident: 10.1016/j.procs.2017.05.286_bib0017
– volume: 11
  issue: 5
  year: 2000
  ident: 10.1016/j.procs.2017.05.286_bib0015
  article-title: Improvements to the SMO Algorithm for SVM Regression
  publication-title: IEEE Transactions on Neural Networks
  doi: 10.1109/72.870050
– volume: 17
  start-page: 642
  issue: 3
  year: 2006
  ident: 10.1016/j.procs.2017.05.286_bib0003
  article-title: Finding Optimal Algorithmic Parameters Using Derivative-Free Optimization
  publication-title: SIAM Journal on Optimization
  doi: 10.1137/040620886
– start-page: 309
  year: 2004
  ident: 10.1016/j.procs.2017.05.286_bib0006
  article-title: Mixed Integer Programming: A Progress Report
– start-page: 257
  year: 2005
  ident: 10.1016/j.procs.2017.05.286_bib0007
  article-title: Cbc User Guide
– volume: 65
  start-page: 83
  year: 2016
  ident: 10.1016/j.procs.2017.05.286_bib0005
  article-title: Feature-based Tuning of Simulated Annealing applied to the Curriculum-based Course Timetabling Problem
  publication-title: Computers & Operations Research
  doi: 10.1016/j.cor.2015.07.002
– start-page: 11
  year: 1997
  ident: 10.1016/j.procs.2017.05.286_bib0002
  article-title: Locally Weighted Learning
  publication-title: Artificial Intelligence Review
  doi: 10.1023/A:1006559212014
SSID ssj0000388917
Score 2.0614574
Snippet Integer Programming (IP) is the most successful technique for solving hard combinatorial optimization problems. Modern IP solvers are very complex programs...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 715
SubjectTerms coin-or branch
cut
data mining
feature based parameter tunning
mixed-integer programming
regression algorithms
Title Data Mining Approach for Feature Based Parameter Tunning for Mixed-Integer Programming Solvers
URI https://dx.doi.org/10.1016/j.procs.2017.05.286
Volume 108
WOSCitedRecordID wos000404959000072&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
journalDatabaseRights – providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 1877-0509
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000388917
  issn: 1877-0509
  databaseCode: M~E
  dateStart: 20100101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV07b9swECbctEOXvoumL3DopsrQm-KYpg8PcRrAbpCpAkVSqAJHDmI7MDL0B_VX9o6kZLUOgrZAF0EmSFPgfeIdT3ffEfJGyVwFMlDouZdwQCm1n3PN_SgTyPeWpKWpsXR8wA4P85MTfjQY_GhzYS5nrGny9Zqf_1dRQxsIG1Nn_0Lc3Z9CA9yD0OEKYofrHwn-vVgKb1xbf4djDDfBhGjt4eeCd6C4FJiOGJaFFInTlSlbZPqM67VWProJMRv4yAZvnRnW7jnGUC_6xqxJMgB8mbh0LA3hOX3anfHnNl1sjGbmauF9GnrH9Ux0ZvwEKxibHiOMO1Fz6LFxkSO_gfUDiUromae0Nxl-3nSAp5HfzlyR54NVfeWNhvvDvhvD5mtar9pWZo3ZiHPGfOSmsXrqmrZ29w7y3v7LbG6oU-XMpmdvaQnrsDhFHSWRsj007K3R75zcRstPcFacNGT40Tjnt8jtiMExDMNEv2_8eciqw02B5-4xW5IrE064Ndf1hlDPuJk-IPfcqYTuWTQ9JAPdPCL324of1CmAx-QrgotacNEWXBSAQx24qAEX7cBFHbhMn1_ARXvgog5cT8iXjx-m-yPfVejwZcyzpR_mioWyUpJzyao0FIlOY6FSeNuFVCKtdCR1Fmqu8iiNmRYCrCGdpSzKSxlJGT8lO8280c8IzcpSxKBwkkpGSSYiXgZxUjHYL-IYBBDukqhdrkI6-nqsojIr2jjF08KscYFrXARpAWu8S952g84te8vN3bNWDoV7YaxhWQBybhr4_F8HviB38Zf16b0kO8uLlX5F7sjLZb24eG0Q9hOhbKye
linkProvider ISSN International Centre
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=Data+Mining+Approach+for+Feature+Based+Parameter+Tunning+for+Mixed-Integer+Programming+Solvers&rft.jtitle=Procedia+computer+science&rft.au=Boas%2C+Matheus+G.+Vilas&rft.au=Santos%2C+Haroldo+G.&rft.au=Martins%2C+Rafael+de+S.O.&rft.au=Merschmann%2C+Luiz+H.C.&rft.date=2017&rft.pub=Elsevier+B.V&rft.issn=1877-0509&rft.eissn=1877-0509&rft.volume=108&rft.spage=715&rft.epage=724&rft_id=info:doi/10.1016%2Fj.procs.2017.05.286&rft.externalDocID=S1877050917309389
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1877-0509&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1877-0509&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1877-0509&client=summon