Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
9657684 | Theoretical Computer Science | 2005 | 20 Pages |
Abstract
An instance of a constraint satisfaction problem is l-consistent if any l constraints of it can be simultaneously satisfied. For a set Î of constraint types, Ïl(Î ) denotes the largest ratio of constraints which can be satisfied in any l-consistent instance composed by constraints of types from Î . In the case of sets Î consisting of finitely many Boolean predicates, we express the limit Ïâ(Î ):=limlââÏl(Î ) as the minimum of a certain functional on a convex set of polynomials. Our results yield a robust deterministic algorithm (for a fixed set Î ) running in time linear in the size of the input and 1/ε which finds either an inconsistent set of constraints (of size bounded by the function of ε) or a truth assignment which satisfies the fraction of at least Ïâ(Î )-ε of the given constraints. We also compute the values of Ïl({P}) for several specific predicates P.
Related Topics
Physical Sciences and Engineering
Computer Science
Computational Theory and Mathematics
Authors
ZdenÄk DvoÅák, Daniel Král', OndÅej Pangrác,