Abstract
A constraint network is arc consistent if any value of its variables is compatible with at least one value of any other variable. The Arc Consistency Problem (ACP) consists in filtering out values of the variables of a given network to obtain one that is arc consistent, without eliminating any solution, i.e. a total value assignment that satisfies all constraints. Enforcing arc consistency, or the more general k-consistency, in a constraint network is a widely used preprocessing step before identifying the set of solutions. ACP is known to be inherently sequential, so in this paper we examine the problem of achieving partial consistency, i.e. filtering out values from the variables so that each value in the new network is compatible with at least one value of not necessarily all, but a constant fraction of the other variables. We call such networks partially arc consistent. We give an algorithm that, for any constraint network, outputs a partially arc consistent subnetwork of it in sublinear (O(√n log n)) parallel time using O(n2) processors. This is the first (to our knowledge) sublinear-time parallel algorithm with polynomially many processors that removes at least a constant fraction of the local inconsistencies of a general constraint network, without eliminating any solution. We also generalize the notion of partiality to the k-consistency problem: We give an algorithm that is faster than the previously known k-consistency algorithms, and which finds a partially k-consistent subnetwork of any given network. Finally, we propose several approximation schemes to a total solution of ACP, and show that they are P-complete. This indicates that the approach of partial solutions, rather than that of approximation schemes, is more promising for parallelism.
This research was partially supported by the European Union ESPRIT Basic Research Projects ALCOM II (contract no. 7141), GEPPCOM (contract no. 9072) and Insight II (contract no. 6019).
This is a preview of subscription content, log in via an institution.
Preview
Unable to display preview. Download preview PDF.
References
Bessiere C.: Arc-consistency and arc-consistency again. Artificial Intelligence 65 (1994) 179–190
Cohen D.A., Cooper M.C., and Jeavons P.G.: Characterizing tractable constraints. Artificial Intelligence 65 (1994) 347–361
Cooper M.C.: An optimal k-consistency algorithm. Artificial Intelligence 41 (1990) 89–95
Cooper P.R. and Swain M.J.: Arc consistency: parallelism and domain dependence. Artificial Intelligence 58 (1992) 207–235
Dechter R.: Constraint networks. In: S. Shapiro (ed.) Encyclopedia of Artificial Intelligence, Wiley, New York, 2nd ed. (1992) 276–285
Dechter R. and Meiri I.: Experimental results of preprocessing algorithms for constraint satisfaction problems. Artificial Intelligence 68 (1994) 211–241
Dechter R. and Pearl J.: Tree clustering for constraint networks. Artificial Intelligence 38 353–366
Dendris N.D., Kalafatis I.A., and Kirousis L.M.: An efficient parallel algortihm for geometrically characterising drawings of a class of 3-D objects. Journal of Mathematical Imaging and Vision 4 (1994) 375–387
Freuder E.C.: Complexity of k-tree structured constraint satisfaction problems. Proceedings of the Eighth National Conference on Artificial Intelligence, Boston, Mass. (1990) 4–9
Goldschlager L.M.: The monotone and planar circuit value problems are log-space complete for P. SIGACT News 9(2) (1977) 25–29
Van Hentenryck P., Deville Y., and Teng C.M.: A generic arc-consistency algorithm and its specializations. Artificial Intelligence 57 (1992) 291–321
Karp R.M. and Ramachandran V.: Parallel algorithms for shared-memory machines. In: J. van Leeuwen (ed.) Handbook of Theoretical Computer Science, Elsevier, Amsterdam (1990)
Kasif S.: On the parallel complexity of discrete relaxation in constraint satisfaction networks. Artificial Intelligence 45 (1990) 275–286
Kasif S. and Delcher A.L.: Local consistency in parallel constraint satisfaction networks. Artificial Intelligence 69 (1994) 307–327
Kirousis L.M.: Fast parallel constraint satisfaction. Artificial Intelligence 64 (1993) 147–160
Lieberherr K.J. and Specker E.: Complexity of partial satisfaction. J. of the ACM 28 (1981) 411–421
Mackworth A.K.: Constraint satisfaction. In: S. Shapiro (ed.) Encyclopedia of Artificial Intelligence, Wiley, New York, 2nd ed. (1992) 285–293
Mackworth A.K. and Freuder E.C.: The complexity of some polynomial network consistency algorithms for constraint satisfaction problems. Artificial Intelligence 25 (1985) 65–74
Mackworth A.K. and Freuder E.C.: The complexity of constraint satisfaction revisited. Artificial Intelligence 59 (1993) 57–62
Samal A. and Henderson T.C.: Parallel consistent labeling algorithms. International Journal of Parallel Programming 16(5) (1987) 341–364
Waltz D.: Understanding line drawings of scenes with shadows. The Psychology of Computer Vision, McGraw-Hill, New York (1975) 19–91
Zhang Y. and Mackworth A.K.: Parallel and distributed algorithms for finite constraint satisfaction problems. Proceedings 3rd IEEE Symposium on Parallel and Distributed Processing, Dallas, TX (1991) 394–397
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Dendris, N.D., Kirousis, L.M., Stamatiou, Y.C., Thilikos, D.M. (1995). Partiality and approximation schemes for local consistency in networks of constraints. In: Thiagarajan, P.S. (eds) Foundations of Software Technology and Theoretical Computer Science. FSTTCS 1995. Lecture Notes in Computer Science, vol 1026. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60692-0_50
Download citation
DOI: https://doi.org/10.1007/3-540-60692-0_50
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60692-5
Online ISBN: 978-3-540-49263-4
eBook Packages: Springer Book Archive