{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Makes three main contributions first we revive the

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: satisfaction, including 2-cnfs, Horn clauses, causal networks, and bounded-width networks. Second, we identify new tractable classes based on the notion of diversity . Third, our empirical tests show that, while on uniform theories directional resolution is indeed ine ective, on problems with special structures, like chains and k-trees, having low w3 , directional resolution greatly outperforms DP-backtracking which is one of the most e ective satis ability algorithm known to date. In conclusion, although directional resolution outperformsy DP-backtracking on some classes of problems, it is not advocated as an e ective method for general satis ability problems. Even when the structure is right, there are other structure-exploiting algorithms, like backjumping, that are likely to be more e ective than BDR-DP in nding a satisfying solution. What we do advocate is that structure-based components should be integrated, together with other heuristics (like unit propagation), into any algorithm that tries to solve satis ability e ectively. At the same time, we have shown that, for some structured domains, directional resolution is an e ective knowledge compilation procedure, compiling knowledge into a form that facilitates ecient model generation and query processing. 29 Acknowledgements We would like to thank Dan Frost, Eddie Schwalb and Rachel Ben-Eliyahu for comments on this paper. Also we would like to thank Dan Frost for running experiments with the backjumping algorithm. 30 References [1] Arnborg, S., Corneil D.G., and Proskurowski A., Complexity of Finding Embedding in a k -tree, SIAM J. Algebraic Discrete Methods 8(2):177-184 (1987). [2] Bertele,U. and Brioschi, F., Nonserial Dynamic Programming, Academic Press, New York, 1972. [3] Ben-Eliyahu, R., and Dechter, R., Default Logic, Propositional Logic and Constraints, in Proceedings of the National Conference on Arti cial Intelligence (AAAI-91), July 1991, Anaheim, CA, pp. 379-385. [4] Crawford, J.M. and Auton, L.D., Experimental Results on the Cross-over Point in Satis ability Problems, in Proceedings of AAAI-93, 1993, pp 21-27. [5] Davis, M., Logemann, G., and Loveland D., A Machine Program for Theorem Proving, Communications of the ACM 5:394-397 (1962). [6] Davis, M. and Putnam, H., A Computing Procedure for Quanti cation Theory, J. ACM 7:201-215 (1960). [7] Dechter, R., and Pearl, J. Network-based Heuristics for Constraint Satisfaction Problems, Arti cial Intelligence 34:1-38 (1987). [8] Dechter, R., and Pearl, J., Directed Constraint Networks: A Relational Framework for Causal Models, in Proceedings of the Twelfth International Joint Conference on Arti cial Intelligence (IJCAI-91), Sidney, Australia, August 1991, pp. 1164-1170. [9] Dechter, R., Enhancement Schemes for Constraint Processing: Backjumping, Learning and Cutset Decomposition, Arti cial Intelligence, 41:273-312 (1990). [10] Even, S., Itai, A., and Shamir, A., On the Complexity of Timetable and MultiCommodity Flow", SIAM Jour...
View Full Document

{[ snackBarMessage ]}