### Citations

1542 | A structural approach to operational semantics
- Plotkin
- 1981
(Show Context)
Citation Context ... in the standard semantics expressions (numerical and Boolean) are always evaluated to ground values, in ursa they may be represented in symbolic propositional form. In terms of operational semantics =-=[59]-=-, 2This is one of the problems from International Mathematical Olympiad (IMO) held in 1992. Mathematical problems from IMOs are typically very challenging problems from different areas of mathematics,... |

1350 | Chaff: Engineering an Efficient SAT Solver. In:
- Moskewicz, Madigan, et al.
- 2001
(Show Context)
Citation Context ...cl (conflict-driven, clause-learning) based extensions of the Davis-PutnamLogemann-Loveland algorithm (dpll) [18, 17, 3]. In recent years, a tremendous advance has been made in sat solving technology =-=[52, 22, 75, 3]-=-. These improvements involve both high-level and low-level algorithmic techniques. The advances in sat solving make possible deciding satisfiability of some industrial sat problems with tens of thousa... |

1261 |
A Computing Procedure For Quantification Theory”,
- Davis, Putnam
- 1960
(Show Context)
Citation Context ... that there is no such assignment. Most of the state-of-the-art complete sat solvers are cdcl (conflict-driven, clause-learning) based extensions of the Davis-PutnamLogemann-Loveland algorithm (dpll) =-=[18, 17, 3]-=-. In recent years, a tremendous advance has been made in sat solving technology [52, 22, 75, 3]. These improvements involve both high-level and low-level algorithmic techniques. The advances in sat so... |

1050 | The complexity of theorem-proving procedures.
- Cook
- 1971
(Show Context)
Citation Context ...l clauses are true. Obviously, sat is a special case of csp, with all variables ranging over the domain {0, 1} and with constraints given as clauses. sat was the first problem shown to be np-complete =-=[15]-=-, and it still holds a central position in the field of computational complexity. Stochastic sat solvers cannot prove the input instance to be unsatisfiable, but may find a solution (i.e., a satisfyin... |

869 | Constraint logic programming: a survey,
- Jaffar, Maher
- 1994
(Show Context)
Citation Context ... approaches for constraint programming including: constraint logic programming over finite domains (clp(fd); combines two declarative programming paradigms – logic programming and constraint solving) =-=[37]-=-, answer set programming (asp; a form of declarative programming with the roots in nonmonotonic reasoning, deductive databases and logic programming with some asp systems using sat solvers) [30], and ... |

728 |
Symbolic execution and program testing
- King
- 1976
(Show Context)
Citation Context ...lly, we present results of a limited experimental comparison between ursa and several other systems. 8.1. Symbolic execution. Operation of ursa is related to symbolic execution. In symbolic execution =-=[40, 57]-=-, program inputs are represented by symbolic values rather than by concrete data and the values of program variables are represented as symbolic expressions. The program is executed by manipulating ex... |

693 |
A machine program for theoremproving..
- Davis, Logemann, et al.
- 1962
(Show Context)
Citation Context ... that there is no such assignment. Most of the state-of-the-art complete sat solvers are cdcl (conflict-driven, clause-learning) based extensions of the Davis-PutnamLogemann-Loveland algorithm (dpll) =-=[18, 17, 3]-=-. In recent years, a tremendous advance has been made in sat solving technology [52, 22, 75, 3]. These improvements involve both high-level and low-level algorithmic techniques. The advances in sat so... |

456 | The DLV system for knowledge representation and reasoning.
- Leone, Pfeifer, et al.
- 2006
(Show Context)
Citation Context ... special-purpose solver. The general problem solving systems include libraries for general purpose programming languages, but also modelling and programming languages built on top of specific solvers =-=[36, 33, 42, 14]-=-. Most modelling languages are highly descriptive and have specific language constructs for certain sorts of constraints. Specific constraints are translated to underlying theories by specific reducti... |

403 |
On the Complexity of Derivation in Propositional Calculus”,
- Tseitin
- 1968
(Show Context)
Citation Context ...l is similar, but it lists values of independent variables in all satisfying valuations. Transforming the formula F that corresponds to the constraint to cnf is performed using Tseitin transformation =-=[70]-=- which is linear in both space and time. The central idea of the transformation is to introduce new (,,definitional“) variables for all subformulae of the input formula, to replace the subformulae wit... |

304 | Vigilante: End-to-end containment of Internet worm epidemics.
- Costa, Crowcroft, et al.
- 2004
(Show Context)
Citation Context ...re verification over three decades ago, and recently it gained a renewed interest. The verification tools using symbolic execution include systems like Java Pathfinder14 [56], Pex15 [68], Vigilante16 =-=[16]-=-. Some of these tools use sat and smt solvers, but they typically handle only machine data-types (and not arbitrary bit-widths). Also, their purpose is generating test suites and finding (single) mode... |

252 |
The OPL optimization programming language.
- Hentenryck
- 1999
(Show Context)
Citation Context ... special-purpose solver. The general problem solving systems include libraries for general purpose programming languages, but also modelling and programming languages built on top of specific solvers =-=[36, 33, 42, 14]-=-. Most modelling languages are highly descriptive and have specific language constructs for certain sorts of constraints. Specific constraints are translated to underlying theories by specific reducti... |

195 | Bounded model checking using satisfiability solving.
- Clarke, Bierea, et al.
- 2001
(Show Context)
Citation Context ... and np-complete problems and in many areas including software and hardware verification, model checking, termination analysis, planning, scheduling, cryptanalysis, electronic design automation, etc. =-=[6, 3, 12, 25, 74, 48, 50]-=-. Typically, translations into sat are performed by specialized, problem-specific tools. However, using a general-purpose system capable of reducing a wide range of problems to sat can simplify this t... |

172 | Compiling constraints in clp(FD),
- Codognet, Diaz
- 1996
(Show Context)
Citation Context ... special-purpose solver. The general problem solving systems include libraries for general purpose programming languages, but also modelling and programming languages built on top of specific solvers =-=[36, 33, 42, 14]-=-. Most modelling languages are highly descriptive and have specific language constructs for certain sorts of constraints. Specific constraints are translated to underlying theories by specific reducti... |

149 | The Quest for Efficient Boolean Satisfiability Solvers
- Zhang, Malik
- 2002
(Show Context)
Citation Context ...cl (conflict-driven, clause-learning) based extensions of the Davis-PutnamLogemann-Loveland algorithm (dpll) [18, 17, 3]. In recent years, a tremendous advance has been made in sat solving technology =-=[52, 22, 75, 3]-=-. These improvements involve both high-level and low-level algorithmic techniques. The advances in sat solving make possible deciding satisfiability of some industrial sat problems with tens of thousa... |

145 | SATzilla: portfolio-based algorithm selection for SAT.
- Xu, Hutter, et al.
- 2008
(Show Context)
Citation Context ... need for transformation to cnf [69]. Choosing among available solvers can be automated by using machine learning techniques for analysis of the generated sat instances (or even input specifications) =-=[73, 55]-=-. For solving optimization problems, instead of the existing naive implementation, we are planning to implement more advanced approaches and to explore the use of MaxSAT and pseudo-Boolean solvers [3]... |

128 | T.: The constrainedness of search.
- Gent, MacIntyre, et al.
- 1996
(Show Context)
Citation Context ...lems), the ratio of the number of clauses and the number of variables in the generated formulae, gets rather stable (as the size of the instance increases) and it reflects the problem constrainedness =-=[29]-=-. dimension 1 2 3 4 5 6 7 8 9 10 11 12 number of solutions 1 0 0 2 10 4 40 92 352 724 2680 14200 number of variables 5 44 115 209 331 480 667 841 1052 1286 1560 1819 number of clauses 5 149 418 794 12... |

122 | MiniZinc: Towards a standard CP modelling language
- Nethercote, Stuckey, et al.
- 2007
(Show Context)
Citation Context ...nces. xcsp is already used in csp competitions as an standard input format.19 A high-lever language MiniZinc (a subset of a language Zinc [47]) also aims at becoming a standard specification language =-=[54]-=-. MiniZinc models can be translated to FlatZinc, a low-level solver input language. There are a number of differences between the languages xcsp, MiniZinc and ursa. xcsp representations are low-level,... |

108 | T.: clasp: a conflict-driven answer set solver. In:
- Gebser, Kaufmann, et al.
- 2007
(Show Context)
Citation Context ...http://argo.matf.bg.ac.rs/downloads.html) is an open-source, flexible, and verified sat solver [45, 44]. 10clasp (http://www.cs.uni-potsdam.de/clasp/) is a solver for (extended) normal logic programs =-=[27]-=-. It combines high-level modelling capacities of asp with state-of-the-art techniques from the area of sat solving and it can be used as an asp solver or a sat solver. clasp was a winner at the asp Co... |

73 |
Pex-white box test generation for .NET.
- Tillmann, Halleux
- 2008
(Show Context)
Citation Context ...roposed for software verification over three decades ago, and recently it gained a renewed interest. The verification tools using symbolic execution include systems like Java Pathfinder14 [56], Pex15 =-=[68]-=-, Vigilante16 [16]. Some of these tools use sat and smt solvers, but they typically handle only machine data-types (and not arbitrary bit-widths). Also, their purpose is generating test suites and fin... |

63 | Answer set programming based on propositional satisfiability.
- Giunchiglia, Lierler, et al.
- 2006
(Show Context)
Citation Context ...language constructs for certain sorts of constraints. Specific constraints are translated to underlying theories by specific reduction techniques. Some modelling systems use sat as the target problem =-=[31, 35]-=- and some of them focus on solving np-complete problems by reduction to sat [10]. In this paper we present a novel approach for solving problems by reducing them to sat. The approach can be seen also ... |

54 | SAT Solving for Termination Analysis with Polynomial Interpretations.
- Fuhs, Giesl, et al.
- 2007
(Show Context)
Citation Context ... and np-complete problems and in many areas including software and hardware verification, model checking, termination analysis, planning, scheduling, cryptanalysis, electronic design automation, etc. =-=[6, 3, 12, 25, 74, 48, 50]-=-. Typically, translations into sat are performed by specialized, problem-specific tools. However, using a general-purpose system capable of reducing a wide range of problems to sat can simplify this t... |

52 | RTL-Datapath Verification using Integer Linear Programming,” in
- Brinkmann, Drechsler
- 2002
(Show Context)
Citation Context ...c.) [2]. In particular, symbolic computations employed by the ursa system are closely related to the theory of bit-vector arithmetic and to decision procedures for this theory based on “bit-blasting” =-=[8, 9]-=-. Since solvers for this theory typically cover all the operators used in ursa, the theory of bit-vector arithmetic can be 34 P. JANIČIĆ used as an underlying theory (instead of propositional logic)... |

51 | Verification of Java programs using symbolic execution and invariant generation.
- Pasareanu, Visser
- 2004
(Show Context)
Citation Context ...n has been proposed for software verification over three decades ago, and recently it gained a renewed interest. The verification tools using symbolic execution include systems like Java Pathfinder14 =-=[56]-=-, Pex15 [68], Vigilante16 [16]. Some of these tools use sat and smt solvers, but they typically handle only machine data-types (and not arbitrary bit-widths). Also, their purpose is generating test su... |

51 |
A survey of new trends in symbolic execution for software testing and analysis.
- Pasareanu, Visser
- 2009
(Show Context)
Citation Context ...lly, we present results of a limited experimental comparison between ursa and several other systems. 8.1. Symbolic execution. Operation of ursa is related to symbolic execution. In symbolic execution =-=[40, 57]-=-, program inputs are represented by symbolic values rather than by concrete data and the values of program variables are represented as symbolic expressions. The program is executed by manipulating ex... |

48 | A framework for representing and solving NP search problems.
- Mitchell, Ternovska
- 2005
(Show Context)
Citation Context ...stem mxg23, focused on np-hard problems, uses a modelling language based on classical first order logic, and for a given specification produces a propositional formula (and passes it to a sat solver) =-=[51, 58]-=-. The system can also translate problem specification to sat extended with cardinality constraints. 21http://bach.istc.kobe-u.ac.jp/sugar/ 22http://users.cecs.anu.edu.au/~jinbo/fzntini/ 23http://www.c... |

48 |
Garćıa de la
- Duck, Stuckey, et al.
- 2004
(Show Context)
Citation Context ...smodels/ 31http://dynadec.com URSA: A SYSTEM FOR UNIFORM REDUCTION TO SAT 27 ilog opl 6.332, a default finite-domain g12/fd solver Mercury (using MiniZinc as its input language) from MiniZinc 1.1.133 =-=[64]-=-, and sat-based systems sugar 1.14.6 and FznTini.34 The ursa system was used with clasp 1.2.0 as an underlying sat solver. We performed experimental comparison between all of the above tools on a prot... |

46 | Solving non-clausal formulas with dpll search.
- Thiffault, Bacchus, et al.
- 2004
(Show Context)
Citation Context ... as the sat competitions show. Within this direction of work, we will also analyze performance of stochastic solvers within ursa. In addition, we are exploring potentials of using non-cnf sat solvers =-=[67, 53, 38]-=- within ursa, which avoids the need for transformation to cnf [69]. Choosing among available solvers can be automated by using machine learning techniques for analysis of the generated sat instances (... |

44 | The smodels system.
- SYRJANEN, NIEMELA
- 2001
(Show Context)
Citation Context ...nguage that combines databases and logic programming) dlv27 [42], asp solvers (used with Smodels specifications and a grounder Lparse 1.1.2) clasp 1.3.228 [27], Cmodels 3.7929 [31] and Smodels 2.3430 =-=[65]-=-, hybrid optimization system (using a custom object-oriented programming language) comet 2.1.131 [49], a hybrid optimization system (using a custom declarative modelling language) ibm 24http://www.smt... |

43 | Compiling problem specifications into SAT
- Cadoli, Schaerf
(Show Context)
Citation Context ...complement representation) that corresponds to the compact encoding. NPSpec is a modelling language for constraint problems, going with a tool Spec2Sat that compiles specifications into sat instances =-=[11]-=-. NPSpec uses a highly declarative style of programming, similar to Datalog (a query language for deductive databases). The semantics of NPSpec is based on the model minimality, an extension of the le... |

43 | SAT-based answer set programming
- Giunchiglia, Lierler, et al.
- 2004
(Show Context)
Citation Context ...ving) [37], answer set programming (asp; a form of declarative programming with the roots in nonmonotonic reasoning, deductive databases and logic programming with some asp systems using sat solvers) =-=[30]-=-, and disjunctive logic programming [42]. There are hybrid systems that use custom specification languages and provide support for constraint programming (e.g., ibm ilog opl, comet, g12). Also, there ... |

41 | Compiling finite linear CSP into SAT.
- Tamura, Taga, et al.
- 2009
(Show Context)
Citation Context ...here are several approaches for encoding csp problems into sat [60]. Probably the most popular basic types of encodings into sat are: the sparse encoding, the compact encoding, and the order encoding =-=[34, 66]-=-. In the sparse encoding, a propositional variable xv,i is defined as true iff the integer variable v has a value i assigned to it. Examples of the sparse encoding are the direct encoding and the supp... |

38 | Propositional satisfiability and constraint programming: A comparative survey
- Bordeaux, Hamadi, et al.
(Show Context)
Citation Context ... and np-complete problems and in many areas including software and hardware verification, model checking, termination analysis, planning, scheduling, cryptanalysis, electronic design automation, etc. =-=[6, 3, 12, 25, 74, 48, 50]-=-. Typically, translations into sat are performed by specialized, problem-specific tools. However, using a general-purpose system capable of reducing a wide range of problems to sat can simplify this t... |

38 | SAT-encodings, search space structure, and local search performance.
- Hoos
- 1999
(Show Context)
Citation Context ...here are several approaches for encoding csp problems into sat [60]. Probably the most popular basic types of encodings into sat are: the sparse encoding, the compact encoding, and the order encoding =-=[34, 66]-=-. In the sparse encoding, a propositional variable xv,i is defined as true iff the integer variable v has a value i assigned to it. Examples of the sparse encoding are the direct encoding and the supp... |

37 |
Flavio Lerda. A Tool for Checking ANSI-C Programs
- Clarke, Kroening
(Show Context)
Citation Context ...n of signed integers or floating point numbers can be, in principle, described and implemented by analogy (http://www.informatik.uni-bremen.de/~florian/sonolar/, http://www.cprover.org/SMT-LIB-Float/ =-=[13, 7]-=-). Still, as discussed by Brillout et al. [7], the bottleneck of a reducing floating point operations to propositional logic would be in the complexity of the resulting propositional formulae. 12 P. J... |

36 | Using auxiliary variables and implied constraints to model non-binary problems
- Smith, Stergiou, et al.
- 2000
(Show Context)
Citation Context ...e for each integer variable v and integer value i [66, 1]. Even within one encoding style, modelling of a problem can take URSA: A SYSTEM FOR UNIFORM REDUCTION TO SAT 25 significantly different forms =-=[63]-=-. There are a number of both theoretical and practical studies and comparisons between different encoding schemes. Since the log encoding lacks the propagation power of the direct and support encoding... |

31 | A comparative study of eight constraint programming languages over the Boolean and finite domains,
- Fernández, Hill
- 2000
(Show Context)
Citation Context ...th all of the above cautious, several (a very few, likely due to the above difficulties) existing reports give some general picture of efficiency of systems for constraint solving over finite domains =-=[24, 43, 21]-=- while new insights are provided by a number of system competitions recently initiated. For experimental comparison with the ursa system, we used the following state-ofthe-art systems (and versions), ... |

31 | Applications of SAT Solvers to Cryptanalysis of Hash Functions,
- Mironov, Zhang
- 2006
(Show Context)
Citation Context |

28 |
Logical cryptanalysis as a SAT problem
- Massacci, Marraro
(Show Context)
Citation Context |

26 | Mixed abstractions for floating-point arithmetic. In:
- Brillout, Kroening, et al.
- 2009
(Show Context)
Citation Context ...n of signed integers or floating point numbers can be, in principle, described and implemented by analogy (http://www.informatik.uni-bremen.de/~florian/sonolar/, http://www.cprover.org/SMT-LIB-Float/ =-=[13, 7]-=-). Still, as discussed by Brillout et al. [7], the bottleneck of a reducing floating point operations to propositional logic would be in the complexity of the resulting propositional formulae. 12 P. J... |

19 | Universal Booleanization of Constraint Models
- Huang
- 2008
(Show Context)
Citation Context ...language constructs for certain sorts of constraints. Specific constraints are translated to underlying theories by specific reduction techniques. Some modelling systems use sat as the target problem =-=[31, 35]-=- and some of them focus on solving np-complete problems by reduction to sat [10]. In this paper we present a novel approach for solving problems by reducing them to sat. The approach can be seen also ... |

19 | Logical loops. In
- Schimpf
- 2002
(Show Context)
Citation Context ...Loop constructs are naturally expressed in ursa, while their absence in some declarative languages (some specification languages do admit complex looping structures) may cause a range of difficulties =-=[62]-=-.20 The ursa language is expressive enough to enable substantially different encodings of the same problem, which is often not easy with other systems. Learning the ursa language should be trivial to ... |

17 | Another look at graph coloring via propositional satisfiability
- Gelder, A
(Show Context)
Citation Context ...isons between different encoding schemes. Since the log encoding lacks the propagation power of the direct and support encodings, it typically leads to less efficient solutions, compared to these two =-=[72, 28]-=-. The order encoding gives better performance compared with the direct encoding and the support encoding for some csp problems [66]. Within the ursa system, numerical variables are represented using b... |

15 |
Garcia de la
- Marriott, Nethercote, et al.
(Show Context)
Citation Context ...y providing a common test-bed of constraint satisfaction instances. xcsp is already used in csp competitions as an standard input format.19 A high-lever language MiniZinc (a subset of a language Zinc =-=[47]-=-) also aims at becoming a standard specification language [54]. MiniZinc models can be translated to FlatZinc, a low-level solver input language. There are a number of differences between the language... |

13 | Xml representation of constraint networks format
- Roussel, Lecoutre
- 2008
(Show Context)
Citation Context ...fferent and incompatible modelling languages. There is no standard modelling language for constraint programming problems. A language xcsp 2.1 is an xmlbased format to represent various csp instances =-=[61]-=-. The main objective of this language is to ease the effort required to test and compare different algorithms by providing a common test-bed of constraint satisfaction instances. xcsp is already used ... |

12 |
On different structure-preserving translations to normal form
- Egly
- 1996
(Show Context)
Citation Context ...onsequently many clauses. There are techniques that can reduce the number of variables and clauses, e.g., by using implications instead of equivalences for subformulae that occur in one polarity only =-=[23]-=-. In the current implementation of the ursa system, for reducing (in some cases) the number of variables and clauses, associativity and commutativity of the connectives ∧ and ∨ is used.7 If a formula ... |

11 | An Abstract Interpretation Based Combinator for Modeling While Loops in Constraint Programming
- Denmat, Gotlieb, et al.
- 2007
(Show Context)
Citation Context ...jack/ 19http://cpai.ucc.ie/09/ 20There are recent ideas for introducing loops in constraint programming in an imperative language style in order to enable prototyping new constraints with less effort =-=[19]-=-. 24 P. JANIČIĆ Example 8.1 illustrates one family of problems that can be simply solved by the ursa system. Specifications of functions in C (analyzed if they are equivalent) can be almost in verba... |

11 | Formalization and implementation of modern sat solvers
- Marić
(Show Context)
Citation Context ...ver would not necessarily take the same time as the solving process can take different routes. 9ArgoSAT (http://argo.matf.bg.ac.rs/downloads.html) is an open-source, flexible, and verified sat solver =-=[45, 44]-=-. 10clasp (http://www.cs.uni-potsdam.de/clasp/) is a solver for (extended) normal logic programs [27]. It combines high-level modelling capacities of asp with state-of-the-art techniques from the area... |

9 |
Efficient sat solving for non-clausal formulas using dpll, graphs, and watched cuts
- Jain, Clarke
- 2009
(Show Context)
Citation Context ... as the sat competitions show. Within this direction of work, we will also analyze performance of stochastic solvers within ursa. In addition, we are exploring potentials of using non-cnf sat solvers =-=[67, 53, 38]-=- within ursa, which avoids the need for transformation to cnf [69]. Choosing among available solvers can be automated by using machine learning techniques for analysis of the generated sat instances (... |

9 | Scala to the power of z3: Integrating smt and programming
- Köksal, Kuncak, et al.
(Show Context)
Citation Context ... and optimisation library with support for several underlying combinatorial solvers, ScalaZ3 is a Scala library for checking satisfiability and solution enumeration with support for the smt solver Z3 =-=[41]-=-. Programs in specification languages used by the modelling systems are generally not directly executed. Rather, they describe a problem at a high-level, descriptive way and the specification does not... |

9 |
CNF encodings.
- Prestwich
- 2009
(Show Context)
Citation Context ...a, a number of different sat solvers, appropriate for different sorts of input problems, following ideas of sat portfolio solvers [55]. There are several approaches for encoding csp problems into sat =-=[60]-=-. Probably the most popular basic types of encodings into sat are: the sparse encoding, the compact encoding, and the order encoding [34, 66]. In the sparse encoding, a propositional variable xv,i is ... |

9 | A sat based scheduler for tournament schedules
- Zhang, Li, et al.
- 2004
(Show Context)
Citation Context |

8 |
as an Effective Solving Technology for Constraint Problems
- SAT
- 2006
(Show Context)
Citation Context ...lated to underlying theories by specific reduction techniques. Some modelling systems use sat as the target problem [31, 35] and some of them focus on solving np-complete problems by reduction to sat =-=[10]-=-. In this paper we present a novel approach for solving problems by reducing them to sat. The approach can be seen also as a general-purpose constraint programming system (for finite domains). The app... |

8 | An empirical study of constraint logic programming and answer set programming solutions of combinatorial problems.
- Dovier, Formisano, et al.
- 2009
(Show Context)
Citation Context ...th all of the above cautious, several (a very few, likely due to the above difficulties) existing reports give some general picture of efficiency of systems for constraint solving over finite domains =-=[24, 43, 21]-=- while new insights are provided by a number of system competitions recently initiated. For experimental comparison with the ursa system, we used the following state-ofthe-art systems (and versions), ... |

7 | A System for Solving Constraint Satisfaction Problems with SMT
- Bofill, Suy, et al.
- 2010
(Show Context)
Citation Context ...ain && bNoCapture); ************************************* ********* URSA Interpreter ********* ************************************* --> Solution 1 n[0]=0 n[1]=6 18 P. JANIČIĆ n[2]=3 n[3]=5 n[4]=7 n=-=[5]-=-=1 n[6]=4 n[7]=2 ... --> Solution 92 n[0]=3 n[1]=1 n[2]=4 n[3]=7 n[4]=5 n[5]=0 n[6]=2 n[7]=6 [Formula generation: 0s; conversion to CNF: 0.01s; total: 0.01s] [Solving time: 0.08s] [Formula size: 841 v... |

7 |
An abstraction-based decision procedure for bit-vector arithmetic.
- Bryant, Kroening, et al.
- 2009
(Show Context)
Citation Context ...c.) [2]. In particular, symbolic computations employed by the ursa system are closely related to the theory of bit-vector arithmetic and to decision procedures for this theory based on “bit-blasting” =-=[8, 9]-=-. Since solvers for this theory typically cover all the operators used in ursa, the theory of bit-vector arithmetic can be 34 P. JANIČIĆ used as an underlying theory (instead of propositional logic)... |

6 | A stochastic Non-CNF SAT solver
- Muhammad, Stuckey
- 2006
(Show Context)
Citation Context ... as the sat competitions show. Within this direction of work, we will also analyze performance of stochastic solvers within ursa. In addition, we are exploring potentials of using non-cnf sat solvers =-=[67, 53, 38]-=- within ursa, which avoids the need for transformation to cnf [69]. Choosing among available solvers can be automated by using machine learning techniques for analysis of the generated sat instances (... |

5 |
The comet programming language and system
- Michel, Hentenryck
- 2005
(Show Context)
Citation Context ...ifications and a grounder Lparse 1.1.2) clasp 1.3.228 [27], Cmodels 3.7929 [31] and Smodels 2.3430 [65], hybrid optimization system (using a custom object-oriented programming language) comet 2.1.131 =-=[49]-=-, a hybrid optimization system (using a custom declarative modelling language) ibm 24http://www.smtlib.org/ 25http://www.probp.com/ 26http://www.sics.se/isl/sicstuswww/site/ 27http://www.dbai.tuwien.a... |

3 |
Satisfiability Modulo Theories, chapter 26, pages 825–885
- Barrett, Sebastiani, et al.
(Show Context)
Citation Context ...j]; } assert_all(bDomain && bNoCapture); ************************************* ********* URSA Interpreter ********* ************************************* --> Solution 1 n[0]=0 n[1]=6 18 P. JANIČIĆ n=-=[2]-=-=3 n[3]=5 n[4]=7 n[5]=1 n[6]=4 n[7]=2 ... --> Solution 92 n[0]=3 n[1]=1 n[2]=4 n[3]=7 n[4]=5 n[5]=0 n[6]=2 n[7]=6 [Formula generation: 0s; conversion to CNF: 0.01s; total: 0.01s] [Solving time: 0.08s]... |

2 |
Predrag Janicic. Instance-based selection of policies for sat solvers
- Nikolic, Maric
- 2009
(Show Context)
Citation Context ...ver in the solving phase. Moreover, it is fruitful to have, within ursa, a number of different sat solvers, appropriate for different sorts of input problems, following ideas of sat portfolio solvers =-=[55]-=-. There are several approaches for encoding csp problems into sat [60]. Probably the most popular basic types of encodings into sat are: the sparse encoding, the compact encoding, and the order encodi... |

1 |
Cabiscol, Inês Lynce, and Felip Manyà. Regular encodings from MAX-CSP into partial MAX-SAT
- Argelich, Alba
- 2009
(Show Context)
Citation Context ...ni] && ni+n[ni]!=nj+n[nj]; } assert_all(bDomain && bNoCapture); ************************************* ********* URSA Interpreter ********* ************************************* --> Solution 1 n[0]=0 n=-=[1]-=-=6 18 P. JANIČIĆ n[2]=3 n[3]=5 n[4]=7 n[5]=1 n[6]=4 n[7]=2 ... --> Solution 92 n[0]=3 n[1]=1 n[2]=4 n[3]=7 n[4]=5 n[5]=0 n[6]=2 n[7]=6 [Formula generation: 0s; conversion to CNF: 0.01s; total: 0.01s... |

1 | Josep Suy, and Mateu Villaret. SIMPLY: a Compiler from a CSP Modeling Language to the SMT-LIB Format - Bofill, Palah́ı - 2009 |

1 |
theory and reflection in theorem proving: a survey and critique
- Meta
- 1995
(Show Context)
Citation Context ...n, its conjuncts are put into cnf separately and conjoined. If any of those conjuncts is a clause, it is directly used (without transformation and definitional variables) at the conjunction top level =-=[32]-=-. The above modifications still keep the transformation linear. 7 The fact that the connectives ∧ and ∨ are associative and commutative is used only when the formula is already generated. An alternati... |

1 |
solver 6.0: Reference manual
- Ilog
- 2003
(Show Context)
Citation Context |

1 |
Predrag Janičić. Formal Correctness Proof for DPLL Procedure
- Marić
(Show Context)
Citation Context ...ver would not necessarily take the same time as the solving process can take different routes. 9ArgoSAT (http://argo.matf.bg.ac.rs/downloads.html) is an open-source, flexible, and verified sat solver =-=[45, 44]-=-. 10clasp (http://www.cs.uni-potsdam.de/clasp/) is a solver for (extended) normal logic programs [27]. It combines high-level modelling capacities of asp with state-of-the-art techniques from the area... |

1 |
Predrag Janičić. Urbiva: Uniform reduction to bit-vector arithmetic
- Marić
- 2010
(Show Context)
Citation Context ...gh-level front-end to the low level smt-lib interchange format, and, further, to all smt solvers that supports it. Reduction to the theory of bit-vector arithmetic is firstly explored in this context =-=[46]-=- and it shows that reducing to bit-vector arithmetic does not necessarily lead to more efficient solving process than reducing to sat (and the same holds for other smt theories). With the increased po... |

1 |
Applications of non-CNF solvers
- Todorović
- 2011
(Show Context)
Citation Context ... analyze performance of stochastic solvers within ursa. In addition, we are exploring potentials of using non-cnf sat solvers [67, 53, 38] within ursa, which avoids the need for transformation to cnf =-=[69]-=-. Choosing among available solvers can be automated by using machine learning techniques for analysis of the generated sat instances (or even input specifications) [73, 55]. For solving optimization p... |

1 |
Logical cryptoanalysis of des
- Šešum
- 2009
(Show Context)
Citation Context ... functions within such framework, propositional formulae corresponding to cryptanalysis tasks were generated (and then solved by a sat solver). Similar approach was also used for cryptanalysis of des =-=[71]-=-. ursa is a general framework aimed not only to cryptanalysis tasks, but to a much wider range of problems. The framework consists of both a modelling language (instead of C++) and the solving machine... |