论文信息 - Symmetry breaking during search in constraint programming

Symmetry breaking during search in constraint programming

We introduce a method for symmetry breaking during search (SBDS) in constraint programming. It has the great advantage of not interfering with heuristic choices. It guarantees to return a unique solution from each set of symmetrically equivalent ones, which is the one found rst by the variable and value ordering heuristics. We prove this claim, describe a general implementation of SBDS in ILOG Solver, and describe applications to low autocorrelation binary sequences and the n-queens problem. We discuss a version of SBDS that can be applied when there are too many symmetries to reason with individually, and give applications in graph colouring and Ramsey theory.

Ian P. Gent | Barbara M. Smith | B. Smith

[1] J. G. Kalbfleisch,et al. On the maximal triangle-free edge-chromatic graphs in three colors , 1968 .

[2] Rolf Backofen,et al. Excluding Symmetries in Concurrent Constraint Programming , 1998 .

[3] Paul Walton Purdom,et al. Backtrack Searching in the Presence of Symmetry , 1988, Nord. J. Comput..

[4] Rolf Backofen. Constraint Techniques for Solving the Protein Structure Prediction Problem , 1998, CP.

[5] James M. Crawford,et al. Symmetry-Breaking Predicates for Search Problems , 1996, KR.

[6] François Pachet,et al. Using Symmetry of Global Constraints to Speed up the Resolution of Constraint Satisfaction Problems , 2002 .

[7] S. Mertens. Exhaustive search for low-autocorrelation binary sequences , 1996 .

[8] Jean-François Puget,et al. On the Satisfiability of Symmetrical Constrained Satisfaction Problems , 1993, ISMIS.