Structure features for SAT instances classification

Abstract The success of portfolio approaches in SAT solving relies on the observation that different SAT solvers may dramatically change their performance depending on the class of SAT instances they are trying to solve. In these approaches, a set of features of the problem is used to build a prediction model, which classifies instances into classes, and computes the fastest algorithm to solve each of them. Therefore, the set of features used to build these classifiers plays a crucial role. Traditionally, portfolio SAT solvers include features about the structure of the problem and its hardness. Recently, there have been some attempts to better characterize the structure of industrial SAT instances. In this paper, we use some structure features of industrial SAT instances to build some classifiers of industrial SAT families of instances. Namely, they are the scale-free structure, the community structure and the self-similar structure. First, we measure the effectiveness of these classifiers by comparing them to other sets of SAT features commonly used in portfolio SAT solving approaches. Then, we evaluate the performance of this set of structure features when used in a real portfolio SAT solver. Finally, we analyze the relevance of these features on the analyzed classifiers.

[1]  C. Sparrow.  The Fractal Geometry of Nature , 1984 .

[2]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[3]  Paul Erdös,et al.  On random graphs, I , 1959 .

[4]  John C. Platt,et al.  Fast training of support vector machines using sequential minimal optimization, advances in kernel methods , 1999 .

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

[6]  Aiko M. Hormann,et al.  Programs for Machine Learning. Part I , 1962, Inf. Control..

[7]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

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

[9]  Oliver Kullmann,et al.  Unified Characterisations of Resolution Hardness Measures , 2014, SAT.

[10]  Albert-László Barabási,et al.  Internet: Diameter of the World-Wide Web , 1999, Nature.

[11]  Daniel Kroening,et al.  Software Verification , 2021, Handbook of Satisfiability.

[12]  William W. Cohen.  Fast Effective Rule Induction , 1995, ICML.

[13]  Maria Luisa Bonet,et al.  On the Classification of Industrial SAT Families , 2015, CCIA.

[14]  David W. Aha,et al.  Instance-Based Learning Algorithms , 1991, Machine Learning.

[15]  Charles Elkan,et al.  Quadratic Programming Feature Selection , 2010, J. Mach. Learn. Res..

[16]  Mark E. J. Newman,et al.  Power-Law Distributions in Empirical Data , 2007, SIAM Rev..

[17]  Predrag Janicic,et al.  Simple algorithm portfolio for SAT , 2011, Artificial Intelligence Review.

[18]  Krzysztof Czarnecki,et al.  SATGraf: Visualizing the Evolution of SAT Formula Structure in Solvers , 2015, SAT.

[19]  Dimitris Achlioptas,et al.  Random Satisfiability , 2009, Handbook of Satisfiability.

[20]  Yong Wang,et al.  Using Model Trees for Classification , 1998, Machine Learning.

[21]  Olivier Roussel,et al.  The International SAT Solver Competitions , 2012, AI Mag..

[22]  Stephen A. Cook,et al.  The complexity of theorem-proving procedures , 1971, STOC.

[23]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[24]  Fuhui Long,et al.  Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy , 2003, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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

[26]  Jussi Rintanen.  Planning and SAT , 2009, Handbook of Satisfiability.

[27]  Maria Luisa Bonet,et al.  Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence (2008) , 2022 .

[28]  John R. Rice,et al.  The Algorithm Selection Problem , 1976, Adv. Comput..

[29]  Kevin Leyton-Brown,et al.  SATzilla: Portfolio-based Algorithm Selection for SAT , 2008, J. Artif. Intell. Res..

[30]  Carlos Ansótegui,et al.  On Balanced CSPs with High Treewidth , 2007, AAAI.

[31]  Yuri Malitsky,et al.  ISAC - Instance-Specific Algorithm Configuration , 2010, ECAI.

[32]  Armin Biere,et al.  Bounded model checking , 2003, Adv. Comput..

[33]  John G. Cleary,et al.  K*: An Instance-based Learner Using and Entropic Distance Measure , 1995, ICML.

[34]  Pat Langley,et al.  Estimating Continuous Distributions in Bayesian Classifiers , 1995, UAI.

[35]  Risto Miikkulainen,et al.  Latent class models for algorithm portfolio methods , 2010, AAAI 2010.

[36]  Yuri Malitsky,et al.  Non-Model-Based Algorithm Portfolios for SAT , 2011, SAT.

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

[38]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[39]  Carlos Ansótegui,et al.  The Impact of Balancing on Problem Hardness in a Highly Structured Domain , 2006, AAAI.

[40]  Predrag Janicic,et al.  Instance-Based Selection of Policies for SAT Solvers , 2009, SAT.

[41]  Jean-Loup Guillaume,et al.  Fast unfolding of communities in large networks , 2008, 0803.0476.

[42]  Bart Selman,et al.  Algorithm portfolios , 2001, Artif. Intell..

[43]  Maria Luisa Bonet,et al.  The Fractal Dimension of SAT Formulas , 2013, IJCAR.

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

[45]  Kevin Leyton-Brown,et al.  : The Design and Analysis of an Algorithm Portfolio for SAT , 2007, CP.

[46]  S. Cessie,et al.  Ridge Estimators in Logistic Regression , 1992 .

[47]  Ulrik Brandes,et al.  On Modularity Clustering , 2008, IEEE Transactions on Knowledge and Data Engineering.