Articles
Ensemble-based prediction of SAT search behaviour

https://doi.org/10.1016/S1571-0653(04)00328-2Get rights and content

Abstract

Abstract

Before attempting to solve an instance of the satisfiability problem, what can we ascertain about the instance at hand and how can we put that information to use when selecting and tuning a SAT algorithm to solve the instance? We argue for an ensemble-based approach and describe an illustrative example of how such a methodology can be applied to determine optimal restart cutoff points for systematic, backtracking search procedures for SAT. We discuss the methodology and indicate how it can be applied to evaluate such strategies as restarts, algorithm comparison, randomization and portfolios of algorithms.

References (30)

  • James M. Crawford et al.
  • Ian P. Gent et al.

    Easy problems are sometimes hard

    Artificial Intelligence

    (1994)
  • Michae. Luby et al.

    Optimal speedup of Las Vegas algorithms

    Info. Proc. Lett.

    (September 1993)
  • Bar. Selman et al.

    Critical behaviour in the computational cost of satisfiability testing in [16]

    (1996)
  • Bar. Selman et al.
  • Roberto J. Bayardo et al.

    Using CSP look-back techniques to solve real-world SAT instances

    Proceedings of the National Conference on Artificial Intelligence

    (1997)
  • P. Cheeseman et al.

    Where the really hard problems are

    Proceedings IHJCAI-91

    (1991)
  • S.A. Cook

    The complexity of theorem proving procedures

    Proceedings of the 3rd Annual ACM Symposium on the Theory of Computation

    (1971)
  • J.M. Crawford et al.

    Experimental results on the application of satisfiability algorithms to scheduling problems

    Proceedings of the National Conference on Artificial Intelligence

    (1994)
  • M. Davis et al.

    A machine program for theorem-proving

    Comm. ACM

    (1962)
  • M. Davis et al.

    A computing procedure for quantification theory

    J. ACM

    (1960)
  • C.P. Gomes et al.

    Algorithm portfolio design: Theory vs. practice

    Proceedings of the Thirteenth Conference on Uncertainty in Artificial Intelligence (UAI-97)

    (1997)
  • Carl. Gomes et al.

    On the fine structure of large search spaces

    IEEE ICTAI'99

    (1999)
  • Carla P. Gomes et al.

    Boosting combinatorial search through randomization

    • Cited by (0)

    1

    Supported by the Spinoza project ‘Logic in Action’ and by a grant from the Netherlands Organization for Scientific Research (NWO) under project number 365-20-005.

    View full text
    Copyright © 2004 Elsevier Ltd. All rights reserved.