Efficient local search for very large-scale satisfiability problems

  • Authors:
  • Jun Gu

  • Affiliations:
  • -

  • Venue:
  • ACM SIGART Bulletin
  • Year:
  • 1992

Quantified Score

Hi-index 0.00

Visualization

Abstract

The satisfiability problem (SAT) is a fundamental problem in mathematical logic, inference, automated reasoning, and computing theory. In this correspondence, we report the results of applying local search techniques to solve the satisfiability problem. While a traditional resolution-based algorithm is not able to handle even moderately sized inference problems, a local search algorithm, tested through years of real program execution, is capable of computing very large-scale inference problems with fast, robust convergence. On a workstation computer, for example, it is able to solve a satisfiability problem with 50,000 clauses and 5,000 variables in a few seconds.