### 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 2^{αn} (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 language | English |
---|---|

Title of host publication | Algorithm Theory — SWAT 1998 - 6th Scandinavian Workshop on Algorithm Theory, Proceedings |

Editors | Stefan Arnborg, Lars Ivansson |

Publisher | Springer Nature |

Pages | 246-254 |

Number of pages | 9 |

ISBN (Print) | 3540646825, 9783540646822 |

DOIs | |

Publication status | Published - 1 Jan 1998 |

Event | 6th Scandinavian Workshop on Algorithm Theory, SWAT 1998 - Stockholm Duration: 8 Jul 1998 → 10 Jul 1998 |

### Publication series

Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|

Volume | 1432 |

ISSN (Print) | 0302-9743 |

ISSN (Electronic) | 1611-3349 |

### Conference

Conference | 6th Scandinavian Workshop on Algorithm Theory, SWAT 1998 |
---|---|

Country | Sweden |

City | Stockholm |

Period | 8/07/98 → 10/07/98 |

### 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