Unformatted text preview: nal on Computing, 5:691703 (1976). 31 [11] Freuder, E.C., A Sucient Condition for BacktrackFree Search, J. ACM, 29:2432 (1982). [12] Galil, Z., On the Complexity of Regular Resolution and the DavisPutnam Procedure, Theoretical Computer Science 4:2346 (1977).
[13] Gashnig, J., Performance Measurement and Analysis of Certain Search Algorithms, Technical Report CMUCS79124, Carnegie Mellon University, 1979. [14] Goerdt, A., DavisPutnam Resolution Versus Unrestricted Resolution, Annals of Math ematics and Articial Intelligence, 6:169184 (1992).
[15] Goldberg, A., Purdom P., and Brown, C., Average Time Analysis of Simplied DavisPutnam Procedures, Information Processing Letters, 15:7275 (1982). [16] McAllester, D., Private communication. [17] Mitchell, D., Selman, B., and Levesque, H., Hard and Easy Distributions of SAT Problems, in Proceedings of AAAI92, 1992, pp. 459465. [18] Seidel, R., A New Method for Solving Constraint Satisfaction Problems, in Proceed ings of the Seventh international joint conference on Articial Intelligence (IJCAI81),
Vancouver, Canada, August 1981, pp. 338342. [19] Selman, B., Levesque H., and Mitchell, D., A New Method for Solving Hard Satisability Problems, in Proceedings of the Tenth National Conference on Articial Intelligence (AAAI92), San Jose, CA, July 1992. [20] Lauritzen, S.L. and Spigelholter, D.J., Local Computations With Probabilities on Graphical Structures and Their Applications to Expert Systems, Journal of the Royal Statistical Society, Series, B, 5:6574 (1988).
[21] van Beek, P. and Dechter, R., On the Minimality and Decomposability of RowConvex Constraint Networks, Journal of the Association of Computing Machinery (JACM), In press, 1995. 32 [22] van Beek, P. and Dechter, R., Constraint Tightness vs Global Consistency, November, 1994. Submitted manuscript. 33 Table 5: BDRDP (bound 3) and DPbacktracking (termination at 20000 deadends) on (k; m)trees Mean values on 50 experiments per each row DPbacktracking
Time: 1st solution Dead ends Time BDR only BDRDP with bound=3
Time: 1st solution Dead ends Number of new clauses DP after BDR DP after BDR 14tree, Nclauses = 11, Ncliques = 100 Total: 401 variable, 1100 clauses 233.2 7475 5.4 17.7 2 298 14tree, Nclauses = 12, Ncliques = 100 Total: 401 variable, 1200 clauses 352.5 10547 7.5 1.2 7 316 14tree, Nclauses = 14, Ncliques = 100 Total: 401 variable, 1400 clauses 14tree, Nclauses = 13, Ncliques = 100 Total: 401 variable, 1300 clauses 328.8 9182 9.8 0.25 3 339 14tree, Nclauses = 14, Ncliques = 100 Total: 401 variable, 1400 clauses 174.2 4551 11.9 0.0 0 329 24tree, Nclauses = 12, Ncliques = 100 Total: 402 variable, 1200 clauses 160.0 4633 6.0 1.6 25 341 24tree, Nclauses = 13, Ncliques = 100 Total: 402 variable, 1300 clauses 95.4 2589 8.3 34 0.1 0 390 Table 6: Query processing on a 3cnf chain (20 subtheories, 5 variables in each) 20 queries of length 1 DP before DR
0.1 0.0 224.2 296.2 240.5 34.1 34.0 0.0 313.0 362.0 0.1 362.8 361.2 0.0 0.7 363.0 68.1 295.5 0.0 367.4 13 1 30811 40841 33234 4753 4754 1 43324 50001 14 50001 50001 1 91 50001 9505 40842 1 50001 DP afterDR
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Time Deadends Time Deadends 35...
View
Full
Document
This note was uploaded on 09/17/2013 for the course PMATH 330 taught by Professor W. alabama during the Spring '09 term at Waterloo.
 Spring '09
 W. Alabama

Click to edit the document details