2 Complexity and Approximability for Parameterized CSPs

The complexity of various Constraint Satisfaction Problems (CSP) when parameterized by structural measures (such as treewidth or clique-width) is a very well-investigated area. In this paper, we take a fresh look at such questions from the point of view of approximation, considering four standard CSP predicates: AND, OR, PARITY, and MAJORITY. By providing new or tighter hardness results for the satisfiability versions, as well as approximation algorithms for the corresponding maximization problems, we show that already these basic predicates display a diverse set of behaviors, ranging from being FPT to optimize exactly for quite general parameters (PARITY), to being W-hard but well-approximable (OR, MAJORITY) to being W-hard and inapproximable (AND). Our results indicate the interest in posing the question of approximability during the usual investigation of CSP complexity with regards to the landscape of structural parameters. 1998 ACM Subject Classification F.2.0

[1]  Thomas J. Schaefer,et al.  The complexity of satisfiability problems , 1978, STOC.

[2]  Stefan Szeider,et al.  Model Counting for Formulas of Bounded Clique-Width , 2013, ISAAC.

[3]  Stefan Szeider.  The Parameterized Complexity of Constraint Satisfaction and Reasoning , 2011, INAP/WLP.

[4]  Saket Saurabh,et al.  Capacitated Domination and Covering: A Parameterized Perspective , 2008, IWPEC.

[5]  Dániel Marx,et al.  Parameterized Complexity and Approximation Algorithms , 2008, Comput. J..

[6]  Edoardo Amaldi,et al.  On the Approximability of Minimizing Nonzero Variables or Unsatisfied Relations in Linear Systems , 1998, Theor. Comput. Sci..

[7]  Daniël Paulusma,et al.  Satisfiability of Acyclic and Almost Acyclic CNF Formulas , 2010, FSTTCS.

[8]  Daniël Paulusma,et al.  Model Counting for CNF Formulas of Bounded Modular Treewidth , 2015, Algorithmica.

[9]  Stefan Szeider,et al.  Kernels for Global Constraints , 2011, IJCAI.

[10]  Prasad Raghavendra,et al.  Agnostic Learning of Monomials by Halfspaces Is Hard , 2009, 2009 50th Annual IEEE Symposium on Foundations of Computer Science.

[11]  Stefan Szeider,et al.  Guarantees and limits of preprocessing in constraint satisfaction and reasoning , 2014, Artif. Intell..

[12]  Elchanan Mossel,et al.  Majority is stablest: discrete and SoS , 2012, STOC '13.

[13]  Martin Grohe.  The Structure of Tractable Constraint Satisfaction Problems , 2006, MFCS.

[14]  Subhash Khot,et al.  Approximating CSPs Using LP Relaxation , 2015, ICALP.

[15]  Luca Trevisan,et al.  Inapproximability of Combinatorial Optimization Problems , 2004, Electron. Colloquium Comput. Complex..

[16]  Stefan Szeider,et al.  Backdoors to Acyclic SAT , 2012, ICALP.

[17]  Khaled M. Elbassioni,et al.  On the approximability of the maximum feasible subsystem problem with 0/1-coefficients , 2009, SODA.

[18]  Johan Håstad,et al.  Some optimal inapproximability results , 2001, JACM.

[19]  Bruno Courcelle,et al.  Linear Time Solvable Optimization Problems on Graphs of Bounded Clique Width , 1998, WG.

[20]  David P. Williamson,et al.  A complete classification of the approximability of maximization problems derived from Boolean constraint satisfaction , 1997, STOC '97.

[21]  Marko Samer,et al.  Constraint satisfaction with bounded treewidth revisited , 2010, J. Comput. Syst. Sci..

[22]  Prasad Raghavendra,et al.  Hardness of Learning Halfspaces with Noise , 2006, FOCS.

[23]  Michael Lampis,et al.  Algorithmic Meta-theorems for Restrictions of Treewidth , 2010, Algorithmica.

[24]  Michael Alekhnovich,et al.  Satisfiability, Branch-Width and Tseitin tautologies , 2002, The 43rd Annual IEEE Symposium on Foundations of Computer Science, 2002. Proceedings..

[25]  Edoardo Amaldi,et al.  The Complexity and Approximability of Finding Maximum Feasible Subsystems of Linear Relations , 1995, Theor. Comput. Sci..

[26]  Subhash Khot,et al.  A characterization of approximation resistance for even k-partite CSPs , 2013, ITCS '13.

[27]  Stefan Szeider,et al.  The parameterized complexity of k-flip local search for SAT and MAX SAT , 2009, Discret. Optim..

[28]  Hendrik W. Lenstra,et al.  Integer Programming with a Fixed Number of Variables , 1983, Math. Oper. Res..

[29]  Egon Wanke,et al.  The Tree-Width of Clique-Width Bounded Graphs Without Kn, n , 2000, WG.

[30]  Jan Arne Telle,et al.  Solving MaxSAT and #SAT on Structured CNF Formulas , 2014, SAT.

[31]  Stefan Szeider,et al.  Not So Easy Problems for Tree Decomposable Graphs , 2011, ArXiv.

[32]  Michael R. Fellows,et al.  Review of: Fundamentals of Parameterized Complexity by Rodney G. Downey and Michael R. Fellows , 2015, SIGA.

[33]  Stefan Rümmele,et al.  Tractable answer-set programming with weight constraints: bounded treewidth is not enough* , 2010, Theory and Practice of Logic Programming.