Cena, Carlo, Akgün, Özgür, Kiziltan, Zeynep et al. (3 more authors) (2023) Learning When to Use Automatic Tabulation in Constraint Model Reformulation. In: Proceedings of the 32nd International Joint Conference on Artificial Intelligence. IJCAI 2023, the 32nd International Joint Conference on Artificial Intelligence, 19-25 Aug 2023 IJCAI/AAAI , MAC , pp. 1902-1910.
Abstract
Combinatorial optimisation has numerous practi- cal applications, such as planning, logistics, or cir- cuit design. Problems such as these can be solved by approaches such as Boolean Satisfiability (SAT) or Constraint Programming (CP). Solver perfor- mance is affected significantly by the model cho- sen to represent a given problem, which has led to the study of model reformulation. One such method is tabulation: rewriting the expression of some of the model constraints in terms of a single table constraint. Successfully applying this process means identifying expressions amenable to trans- formation, which has typically been done manu- ally. Recent work introduced an automatic tab- ulation using a set of hand-designed heuristics to identify constraints to tabulate. However, the per- formance of these heuristics varies across problem classes and solvers. Recent work has shown learn- ing techniques to be increasingly useful in the con- text of automatic model reformulation. The goal of this study is to understand whether it is possi- ble to improve the performance of such heuristics, by learning a model to predict whether or not to activate them for a given instance. Experimental results suggest that a random forest classifier is the most robust choice, improving the performance of four different SAT and CP solvers.
Metadata
Item Type: | Proceedings Paper | ||||
---|---|---|---|---|---|
Authors/Creators: |
|
||||
Copyright, Publisher and Additional Information: | This is an author-produced version of the published paper. Uploaded in accordance with the University’s Research Publications and Open Access policy. |
||||
Dates: |
|
||||
Institution: | The University of York | ||||
Academic Units: | The University of York > Faculty of Sciences (York) > Computer Science (York) | ||||
Funding Information: |
|
||||
Depositing User: | Pure (York) | ||||
Date Deposited: | 20 Jul 2023 10:30 | ||||
Last Modified: | 22 Apr 2024 23:05 | ||||
Published Version: | https://doi.org/10.24963/ijcai.2023/211 | ||||
Status: | Published | ||||
Publisher: | IJCAI/AAAI | ||||
Refereed: | No | ||||
Identification Number: | https://doi.org/10.24963/ijcai.2023/211 |