Options
noSAT-MaxSAT
Citation Link: https://doi.org/10.15480/882.4769
Publikationstyp
Conference Paper
Date Issued
2022
Sprache
English
Author(s)
Institut
TORE-DOI
Number in series
B-2022-2
Start Page
29
End Page
30
Citation
Bacchus, F., Berg, J., Järvisalo, M., Martins, R., & Niskanen, A. (Eds.) (2022). MaxSAT Evaluation 2022: Solver and Benchmark Descriptions. (Department of Computer Science Series of Publications B; Vol. B-2022-2). Department of Computer Science, University of Helsinki. - Seite 29-30 (2022)
Contribution to Conference
Publisher
Department of Computer Science, University of Helsinki
Peer Reviewed
false
Since 2019, all solvers in the incomplete track of the MaxSAT Evaluation (MSE) include a dedicated SAT solver. In resource-constrained computing environments, e.g., embedded systems, such algorithm designs can be hard to realize. Yet, the MSE results show an undeniable increase in efficiency when using a SAT solver. With noSAT-MaxSAT, we propose to replace the call to a SAT solver by executing the MaxSAT solver only on the hard clauses of the formula instead. The algorithm is based on SATLike, which was was the only algorithm that did not rely on a SAT solver in the MaxSAT Evaluation 2018. Additionally, our implementation satisfies a set of requirements often found in embedded system software.
Subjects
MaxSAT
DDC Class
004: Informatik
Publication version
publishedVersion
Loading...
Name
noSAT-MaxSAT.pdf
Size
107.44 KB
Format
Adobe PDF