Linear-programming design and analysis of fast algorithms for Max 2-CSP

The class Max (r,2)-CSP, or simply Max 2-CSP, consists of constraint satisfaction problems with at most two r-valued variables per clause. For instances with n variables and m binary clauses, we present an O(nr^5^+^1^9^m^/^1^0^0)-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^(^1^3^/^7^5^+^o^(^1^)^)^m), this is fastest for most problems in Max 2-CSP. Parametrizing in terms of n rather than m, for graphs of average degree d we show a simple algorithm running time O^@?(2^(^1^-^2^d^+^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.

[1]  DechterRina,et al.  Tree clustering for constraint networks (research note) , 1989 .

[2]  David Eppstein,et al.  3-Coloring in Time O(1.3289^n) , 2000, J. Algorithms.

[3]  Luca Trevisan,et al.  Gadgets, Approximation, and Linear Programming , 2000, SIAM J. Comput..

[4]  Ryan Williams,et al.  A new algorithm for optimal 2-constraint satisfaction and its implications , 2005, Theor. Comput. Sci..

[5]  Rolf Niedermeier,et al.  Worst-case upper bounds for MAX-2-SAT with an application to MAX-CUT , 2003, Discret. Appl. Math..

[6]  Nadia Creignou,et al.  A Dichotomy Theorem for Maximum Generalized Satisfiability Problems , 1995, J. Comput. Syst. Sci..

[7]  Rina Dechter,et al.  Tree Clustering for Constraint Networks , 1989, Artif. Intell..

[8]  Alex D. Scott,et al.  Solving Sparse Random Instances of Max Cut and Max 2-CSP in Linear Expected Time , 2006, Combinatorics, Probability and Computing.

[9]  Luca Trevisan,et al.  The Approximability of Constraint Satisfaction Problems , 2001, SIAM J. Comput..

[10]  A. E. Eiben,et al.  Constraint-satisfaction problems. , 2000 .

[11]  Peter Jonsson,et al.  An algorithm for counting maximum weighted independent sets and its applications , 2002, SODA '02.

[12]  Alex D. Scott,et al.  An LP-Designed Algorithm for Constraint Satisfaction , 2006, ESA.

[13]  Edward A. Hirsch,et al.  A New Algorithm for MAX-2-SAT , 2000, STACS.

[14]  U. Schöning.  A probabilistic algorithm for k-SAT and constraint satisfaction problems , 1999, 40th Annual Symposium on Foundations of Computer Science (Cat. No.99CB37039).

[15]  Magnus Wahlström,et al.  Counting models for 2SAT and 3SAT formulae , 2005, Theor. Comput. Sci..

[16]  Alexander S. Kulikov,et al.  A new approach to proving upper bounds for MAX-2-SAT , 2006, SODA '06.

[17]  P. Rossmanith,et al.  A New Satisabilit y Algorithm With Applications To Max-Cut , 2005 .

[18]  David Eppstein,et al.  Quasiconvex analysis of backtracking algorithms , 2003, SODA '04.

[19]  EppsteinDavid,et al.  3-coloring in time O(1.3289n) , 2005 .

[20]  Noga Alon,et al.  Large induced degenerate subgraphs , 1987, Graphs Comb..

[21]  Gregory B. Sorkin,et al.  Generalized Constraint Satisfaction Problems , 2006 .

[22]  Fabrizio Grandoni,et al.  Some New Techniques in Design and Analysis of Exact (Exponential) Algorithms , 2005, Bull. EATCS.

[23]  Burkhard Monien,et al.  Upper bounds on the bisection width of 3- and 4-regular graphs , 2006, J. Discrete Algorithms.

[24]  F. Della Croce,et al.  An exact algorithm for MAX-CUT in sparse graphs , 2007, Oper. Res. Lett..

[25]  Klaus Jansen,et al.  Polynomial Time Approximation Schemes for MAX-BISECTION on Planar and Geometric Graphs , 2001, SIAM J. Comput..

[26]  Stefan Richter,et al.  Algorithms Based on the Treewidth of Sparse Graphs , 2005, WG.

[27]  Alex D. Scott,et al.  Faster Algorithms for MAX CUT and MAX CSP, with Polynomial Expected Time for Sparse Instances , 2003, RANDOM-APPROX.

[28]  Rina Dechter,et al.  Network-Based Heuristics for Constraint-Satisfaction Problems , 1987, Artif. Intell..

[29]  Fedor V. Fomin,et al.  Reports in Informatics , 2005 .

[30]  Rolf Niedermeier,et al.  New Upper Bounds for Maximum Satisfiability , 2000, J. Algorithms.

[31]  Stefan Arnborg,et al.  Linear time algorithms for NP-hard problems restricted to partial k-trees , 1989, Discret. Appl. Math..

[32]  Dániel Marx.  Parameterized complexity of constraint satisfaction problems , 2004 .

[33]  Martin Fürer,et al.  Algorithms for Counting 2-SAT Solutions and Colorings with Applications , 2005, Electron. Colloquium Comput. Complex..

[34]  Efficient Algorithms and Partial k-trees 1 , .

[35]  Ryan Williams.  A new algorithm for optimal constraint satisfaction and its implications , 2004, Electron. Colloquium Comput. Complex..

[36]  Alex D. Scott,et al.  Polynomial Constraint Satisfaction: A Framework for Counting and Sampling CSPs and Other Problems , 2006, ArXiv.

[37]  Klaus Jansen,et al.  Polynomial Time Approximation Schemes for MAX-BISECTION on Planar and Geometric Graphs , 2005, SIAM J. Comput..

[38]  Martin Fürer,et al.  Exact Max 2-Sat: Easier and Faster , 2007, SOFSEM.

[39]  Uwe Schöning.  A Probabilistic Algorithm for k-SAT and Constraint Satisfaction Problems , 1999, FOCS.