A modular approach to MaxSAT modulo theories

Authors:
Alessandro Cimatti;Alberto Griggio;Bastiaan Joost Schaafsma;Roberto Sebastiani
Affiliations:
FBK-IRST, Trento, Italy;FBK-IRST, Trento, Italy;FBK-IRST, Trento, Italy,DISI, University of Trento, Italy;DISI, University of Trento, Italy
Venue:
SAT'13 Proceedings of the 16th international conference on Theory and Applications of Satisfiability Testing
Year:
2013

Citing 15
Cited 1

Efficient theory combination via boolean search

Information and Computation - Special issue: Combining logical systems
New inference rules for Max-SAT

Journal of Artificial Intelligence Research
MINIMAXSAT: an efficient weighted max-SAT solver

Journal of Artificial Intelligence Research
Z3: an efficient SMT solver

TACAS'08/ETAPS'08 Proceedings of the Theory and practice of software, 14th international conference on Tools and algorithms for the construction and analysis of systems
Computing small unsatisfiable cores in satisfiability modulo theories

Journal of Artificial Intelligence Research
Solving MAXSAT by solving a sequence of simpler SAT instances

CP'11 Proceedings of the 17th international conference on Principles and practice of constraint programming
Splitting on demand in SAT modulo theories

LPAR'06 Proceedings of the 13th international conference on Logic for Programming, Artificial Intelligence, and Reasoning
On SAT modulo theories and optimization problems

SAT'06 Proceedings of the 9th international conference on Theory and Applications of Satisfiability Testing
On solving the partial MAX-SAT problem

SAT'06 Proceedings of the 9th international conference on Theory and Applications of Satisfiability Testing
Satisfiability modulo the theory of costs: foundations and applications

TACAS'10 Proceedings of the 16th international conference on Tools and Algorithms for the Construction and Analysis of Systems
Improvements to core-guided binary search for MaxSAT

SAT'12 Proceedings of the 15th international conference on Theory and Applications of Satisfiability Testing
Optimization in SMT with LA(Q) cost functions

IJCAR'12 Proceedings of the 6th international joint conference on Automated Reasoning
SAT-based MaxSAT algorithms

Artificial Intelligence
The MathSAT5 SMT solver

TACAS'13 Proceedings of the 19th international conference on Tools and Algorithms for the Construction and Analysis of Systems
Solving weighted CSPs with meta-constraints by reformulation into satisfiability modulo theories

Constraints

Iterative and core-guided MaxSAT solving: A survey and assessment

Constraints

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a novel "modular" approach for (weighted partial) MaxSAT Modulo Theories. The main idea is to combine a lazy SMT solver with a purely-propositional (weighted partial) MaxSAT solver, by making them exchange information iteratively: the former produces an increasing set of theory lemmas which are used by the latter to progressively refine an approximation of the final subset of the soft clauses, which is eventually returned as output. The approach has several practical features. First, it is independent from the theories addressed. Second, it is simple to implement and to update, since both SMT and MaxSAT solvers can be used as blackboxes. Third, it can be interfaced with external MaxSAT and SMT solvers in a plug-and-play manner, so that to benefit for free of tools which are or will be made available. We have implemented our approach on top of the MathSAT5 SMT solver and of a selection of external MaxSAT solvers, and we have evaluated it by means of an extensive empirical test on SMT-LIB benchmarks. The results confirm the validity and potential of this approach.