Article ID Journal Published Year Pages File Type
1141601 Discrete Optimization 2007 28 Pages PDF
Abstract

The class Max (r,2)(r,2)-CSP, or simply Max 2-CSP, consists of constraint satisfaction problems with at most two rr-valued variables per clause. For instances with nn variables and mm binary clauses, we present an O(nr5+19m/100)O(nr5+19m/100)-time algorithm which is the fastest polynomial-space algorithm for many problems in the class, including Max Cut. The method also proves a treewidth bound tw(G)≤(13/75+o(1))m, which gives a faster Max 2-CSP algorithm that uses exponential space: running in time O⋆(2(13/75+o(1))m)O⋆(2(13/75+o(1))m), this is fastest for most problems in Max 2-CSP. Parametrizing in terms of nn rather than mm, for graphs of average degree dd we show a simple algorithm running time O⋆(2(1−2d+1)n), the fastest polynomial-space algorithm known.In combination with “Polynomial CSPs” introduced in a companion paper, these algorithms also allow (with an additional polynomial factor overhead in space and time) counting and sampling, and the solution of problems like Max Bisection that escape the usual CSP framework.Linear programming is key to the design as well as the analysis of the algorithms.

Related Topics
Physical Sciences and Engineering Mathematics Control and Optimization
Authors
, ,