Skip to Main content Skip to Navigation
Conference papers

Sur l'UP-résilience des k-UCSs binaires

Abstract : Abstract Branch and Bound (BnB) solvers for Max-SAT exploit max-resolution, the inference rule for Max-SAT, to ensure that every computed Inconsistent Subset (IS) is counted only once. However, learning max-resolution transformations can be detrimental to their performance so they are usually selectively learned if they respect certain patterns. In this paper, we focus on particular patterns called binary k-UCSs. We prove that these patterns verify a recent characterization of max-resolution transformations called UP-resilience and we show how this result can help extend the current patterns. Finally, this work is part of a global approach to characterize the relevance of transformations by max-resolution.
Document type :
Conference papers
Complete list of metadatas

Cited literature [8 references]  Display  Hide  Download

https://hal-amu.archives-ouvertes.fr/hal-02295161
Contributor : Djamal Habet <>
Submitted on : Tuesday, September 24, 2019 - 8:46:58 AM
Last modification on : Sunday, December 15, 2019 - 3:22:03 PM
Long-term archiving on: : Sunday, February 9, 2020 - 7:24:29 AM

File

ArticleJFPC2019.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02295161, version 1

Collections

Citation

Mohamed Sami Cherif, Djamal Habet. Sur l'UP-résilience des k-UCSs binaires. Actes des 15èmes Journées Francophones de Programmation par Contraintes (JFPC), pp. 177-180, Jun 2019, Albi, France. ⟨hal-02295161⟩

Share

Metrics

Record views

77

Files downloads

23