The Impact of Heterogeneity and Geometry on the Proof Complexity of Random Satisfiability

Satisfiability is considered the canonical NP-complete problem and is used as a starting point for hardness reductions in theory, while in practice heuristic SAT solving algorithms can solve large-scale industrial SAT instances very efficiently. This disparity between theory and practice is believed to be a result of inherent properties of industrial SAT instances that make them tractable. Two characteristic properties seem to be prevalent in the majority of real-world SAT instances, heterogeneous degree distribution and locality. To understand the impact of these two properties on SAT, we study the proof complexity of random k-SAT models that allow to control heterogeneity and locality. Our findings show that heterogeneity alone does not make SAT easy as heterogeneous random k-SAT instances have superpolynomial resolution size. This implies intractability of these instances for modern SAT-solvers. On the other hand, modeling locality with an underlying geometry leads to small unsatisfiable subformulas, which can be found within polynomial time. A key ingredient for the result on geometric random k-SAT can be found in the complexity of higher-order Voronoi diagrams. As an additional technical contribution, we show a linear upper bound on the number of non-empty Voronoi regions, that holds for points with random positions in a very general setting. In particular, it covers arbitrary p-norms, higher dimensions, and weights affecting the area of influence of each point multiplicatively. This is in stark contrast to quadratic lower bounds for the worst case.

[1]  Thomas Bläsius,et al.  Efficient Shortest Paths in Scale-Free Networks with Underlying Hyperbolic Geometry , 2018, ICALP.

[2]  Raimund Seidel,et al.  On the number of faces in higher-dimensional Voronoi diagrams , 1987, SCG '87.

[3]  Endre Szemerédi,et al.  Many hard examples for resolution , 1988, JACM.

[4]  Marián Boguñá,et al.  Popularity versus similarity in growing networks , 2011, Nature.

[5]  Ketan Mulmuley,et al.  On levels in arrangements and voronoi diagrams , 1991, Discret. Comput. Geom..

[6]  Friedhelm Meyer auf der Heide,et al.  Average case complexity of Voronoi diagrams of n sites from the unit cube , 2005, EuroCG.

[7]  F. Chung,et al.  Connected Components in Random Graphs with Given Expected Degree Sequences , 2002 .

[8]  Mariette Yvinec,et al.  Voronoi Diagrams in Higher Dimensions under Certain Polyhedral Distance Functions , 1995, SCG '95.

[9]  Michael Ian Shamos,et al.  Closest-point problems , 1975, 16th Annual Symposium on Foundations of Computer Science (sfcs 1975).

[10]  Nikolaos Fountoulakis,et al.  On the giant component of random hyperbolic graphs , 2013 .

[11]  Krzysztof Czarnecki,et al.  The Effect of Structural Measures and Merges on SAT Solver Performance , 2018, CP.

[12]  Maria Luisa Bonet,et al.  Towards Industrial-Like Random SAT Instances , 2009, IJCAI.

[13]  D. T. Lee,et al.  On k-Nearest Neighbor Voronoi Diagrams in the Plane , 1982, IEEE Transactions on Computers.

[14]  Dmitri V. Krioukov,et al.  Scale-free Networks Well Done , 2018, Physical Review Research.

[15]  F. Chung,et al.  The average distances in random graphs with given expected degrees , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[16]  Krzysztof Czarnecki,et al.  Relating Complexity-theoretic Parameters with SAT Solver Performance , 2017, ArXiv.

[17]  Ralph Keusch,et al.  Sampling Geometric Inhomogeneous Random Graphs in Linear Time , 2017, ESA.

[18]  Jesús Giráldez-Cru,et al.  A Modularity-Based Random SAT Instances Generator , 2015, IJCAI.

[19]  Andrew M. Sutton,et al.  On the Empirical Time Complexity of Scale-Free 3-SAT at the Phase Transition , 2019, TACAS.

[20]  Jesús Giráldez-Cru,et al.  Locality in Random SAT Instances , 2017, IJCAI.

[21]  Tobias Müller,et al.  The diameter of KPKVB random graphs , 2019, Advances in Applied Probability.

[22]  Carlos Ansótegui,et al.  The Community Structure of SAT Formulas , 2012, SAT.

[23]  Eli Ben-Sasson,et al.  Space complexity of random formulae in resolution , 2003, Random Struct. Algorithms.

[24]  Adnan Darwiche,et al.  On the power of clause-learning SAT solvers as resolution engines , 2011, Artif. Intell..

[25]  Cecilia Bohler,et al.  On the Complexity of Higher Order Abstract Voronoi Diagrams , 2013, ICALP.

[26]  B. Boots.  Weighting Thiessen Polygons , 1980 .

[27]  D. T. Lee,et al.  An Output-Sensitive Approach for the L 1/L ∞ k-Nearest-Neighbor Voronoi Diagram , 2011, ESA.

[28]  Martin Raab,et al.  "Balls into Bins" - A Simple and Tight Analysis , 1998, RANDOM.

[29]  D. T. Lee,et al.  Higher Order City Voronoi Diagrams , 2012, SWAT.

[30]  Jacobo Torán,et al.  Space Bounds for Resolution , 1999, STACS.

[31]  Sariel Har-Peled,et al.  On the Complexity of Randomly Weighted Multiplicative Voronoi Diagrams , 2015, Discret. Comput. Geom..

[32]  Andrew M. Sutton,et al.  Phase Transitions for Scale-Free SAT Formulas , 2017, AAAI.

[33]  Amin Vahdat,et al.  Hyperbolic Geometry of Complex Networks , 2010, Physical review. E, Statistical, nonlinear, and soft matter physics.

[34]  Rex A. Dwyer.  Higher-dimensional voronoi diagrams in linear expected time , 1989, SCG '89.

[35]  Tobias Friedrich,et al.  Solving Vertex Cover in Polynomial Time on Hyperbolic Random Graphs , 2019, Theory of Computing Systems.

[36]  Eli Ben-Sasson,et al.  Short proofs are narrow—resolution made simple , 2001, JACM.

[37]  Jeff Erickson,et al.  Nice Point Sets Can Have Nasty Delaunay Triangulations , 2001, SCG '01.

[38]  Ashish Sabharwal,et al.  Non-Restarting SAT Solvers with Simple Preprocessing Can Efficiently Simulate Resolution , 2014, AAAI.

[39]  Tobias Friedrich,et al.  The Satisfiability Threshold for Non-Uniform Random 2-SAT , 2019, ICALP.

[40]  Maria Luisa Bonet,et al.  On the Structure of Industrial SAT Instances , 2009, CP.

[41]  Hyeon-Suk Na,et al.  On the average complexity of 3D-Voronoi diagrams of random points on convex polytopes , 2003, Comput. Geom..

[42]  Franz Aurenhammer,et al.  An optimal algorithm for constructing the weighted voronoi diagram in the plane , 1984, Pattern Recognit..

[43]  Sariel Har-Peled,et al.  On the Expected Complexity of Voronoi Diagrams on Terrains , 2016, TALG.

[44]  Sebastian Fischmeister,et al.  Impact of Community Structure on SAT Solver Performance , 2014, SAT.

[45]  V. Klee.  On the complexity ofd- dimensional Voronoi diagrams , 1979 .

[46]  Sanjit A. Seshia,et al.  On the Hardness of SAT with Community Structure , 2016, SAT.

[47]  Jeff Erickson,et al.  Dense Point Sets Have Sparse Delaunay Triangulations or “... But Not Too Nasty” , 2001, SODA '02.

[48]  Alessandro Panconesi,et al.  Concentration of Measure for the Analysis of Randomized Algorithms , 2009 .

[49]  Thomas Sauerwald,et al.  Bounds on the Satisfiability Threshold for Power Law Distributed Random SAT , 2017, ESA.

[50]  Jan Vitek,et al.  Redundancy and coverage detection in sensor networks , 2006, TOSN.

[51]  Jeff Erickson,et al.  Dense point sets have sparse Delaunay triangulations , 2001, ArXiv.

[52]  Antonio Galvão Naclério Novaes,et al.  A multiplicatively-weighted Voronoi diagram approach to logistics districting , 2006, Comput. Oper. Res..