Adaptive Singleton-based Consistencies

Abstract : Singleton-based consistencies have been shown to dra- matically improve the performance of constraint solvers on some difficult instances. However, they are in gen- eral too expensive to be applied exhaustively during the whole search. In this paper, we focus on partition-one- AC, a singleton-based consistency which, as opposed to singleton arc consistency, is able to prune values on all variables when it performs singleton tests on one of them. We propose adaptive variants of partition-one- AC that do not necessarily run until having proved the fixpoint. The pruning can be weaker than the full ver- sion but the computational effort can be significantly reduced. Our experiments show that adaptive Partition- one-AC can obtain significant speed-ups over arc con- sistency and over the full version of partition-one-AC.
Document type :
Conference papers
Complete list of metadatas

Cited literature [9 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01067213
Contributor : Joël Quinqueton <>
Submitted on : Tuesday, February 14, 2017 - 5:15:16 PM
Last modification on : Monday, July 16, 2018 - 3:42:03 PM
Long-term archiving on : Monday, May 15, 2017 - 5:51:56 PM

File

apoac_aaai_2014.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : lirmm-01067213, version 1

Collections

Citation

Amine Balafrej, Christian Bessière, Gilles Trombettoni, El Houssine Bouyakhf. Adaptive Singleton-based Consistencies. AAAI Conference on Artificial Intelligence, Jul 2014, Québec City, Canada. pp.2601-2607. ⟨lirmm-01067213⟩

Share

Metrics

Record views

158

Files downloads

253