Learning when to use a decomposition

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Applying a Dantzig-Wolfe decomposition to a mixed-integer program (MIP) aims at exploiting an embedded model structure and can lead to significantly stronger reformulations of the MIP. Recently, automating the process and embedding it in standard MIP solvers have been proposed, with the detection of a decomposable model structure as key element. If the detected structure reflects the (usually unknown) actual structure of the MIP well, the solver may be much faster on the reformulated model than on the original. Otherwise, the solver may completely fail. We propose a supervised learning approach to decide whether or not a reformulation should be applied, and which decomposition to choose when several are possible. Preliminary experiments with a MIP solver equipped with this knowledge show a significant performance improvement on structured instances, with little deterioration on others.

Original languageEnglish
Title of host publicationIntegration of AI and OR Techniques in Constraint Programming - 14th International Conference, CPAIOR 2017, Proceedings
EditorsDomenico Salvagnin, Michele Lombardi
PublisherSpringer Verlag
Pages202-210
Number of pages9
ISBN (Print)9783319597751
DOIs
Publication statusPublished - 1 Jan 2017
Externally publishedYes
Event14th International Conference on Integration of Artificial Intelligence and Operations Research Techniques in Constraint Programming, CPAIOR 2017 - Padova, Italy
Duration: 5 Jun 20178 Jun 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10335 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Conference on Integration of Artificial Intelligence and Operations Research Techniques in Constraint Programming, CPAIOR 2017
Country/TerritoryItaly
CityPadova
Period5/06/178/06/17

Keywords

  • Automatic Dantzig-Wolfe decomposition
  • Branch-and-price
  • Column generation
  • Mixed-integer programming
  • Supervised learning

Fingerprint

Dive into the research topics of 'Learning when to use a decomposition'. Together they form a unique fingerprint.

Cite this