Local search algorithms for SAT: Worst-case analysis

Research outputpeer-review

2 Citations (Scopus)

Abstract

Recent experiments demonstrated that local search algorithms (e.g. GSAT) are able to find satisfying assignments for many "hard" Boolean formulas. However, no non-trivial worst-case upper bounds were proved, although many such bounds of the form 2αn (α < 1 is a constant) are known for other SAT algorithms, e.g. resolution-like algorithms. In the present paper we prove such a bound for a local search algorithm, namely for CSAT. The class of formulas we consider covers most of DIMACS benchmarks, the satisfiability problem for this class of formulas is NP-complete.

Original languageEnglish
Title of host publicationAlgorithm Theory — SWAT 1998 - 6th Scandinavian Workshop on Algorithm Theory, Proceedings
EditorsStefan Arnborg, Lars Ivansson
PublisherSpringer Nature
Pages246-254
Number of pages9
ISBN (Print)3540646825, 9783540646822
DOIs
Publication statusPublished - 1 Jan 1998
Event6th Scandinavian Workshop on Algorithm Theory, SWAT 1998 - Stockholm
Duration: 8 Jul 199810 Jul 1998

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1432
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference6th Scandinavian Workshop on Algorithm Theory, SWAT 1998
CountrySweden
CityStockholm
Period8/07/9810/07/98

Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Local search algorithms for SAT: Worst-case analysis'. Together they form a unique fingerprint.

  • Cite this

    Hirsch, E. A. (1998). Local search algorithms for SAT: Worst-case analysis. In S. Arnborg, & L. Ivansson (Eds.), Algorithm Theory — SWAT 1998 - 6th Scandinavian Workshop on Algorithm Theory, Proceedings (pp. 246-254). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1432). Springer Nature. https://doi.org/10.1007/BFb0054372