Constraint satisfaction problems in clausal form
نویسنده
چکیده
This is the report-version of a mini-series of two articles [60, 61] on the foundations of conjunctive normal forms with non-boolean variables. These two parts are here bundled in one report, each part yielding a chapter. Part I We consider the problem of generalising boolean formulas in conjunctive normal form by allowing non-boolean variables, with the goal of maintaining combinatorial properties. Requiring that a literal involves only a single variable, the most general form of literals are the well-known “signed literals”, corresponding to unary constraints in CSP. However we argue that only the restricted form of “negative monosigned literals” and the resulting generalised clause-sets, corresponding to “sets of no-goods” in the AI literature, maintain the essential properties of boolean conjunctive normal forms. In this first part of a mini-series of two articles, we build up a solid foundation for (generalised) clause-sets, including the notion of autarky systems, the interplay between autarkies and resolution, and basic notions of (DP-)reductions. As a basic combinatorial parameter of generalised clause-sets we introduce the (generalised) notion of deficiency, which in the boolean case is the difference between the number of clauses and the number of variables. Autarky theory plays a fundamental role here, and we concentrate especially on matching autarkies (based on matching theory). A natural task is to determine the structure of (matching) lean clause-sets, which do not admit non-trivial (matching) autarkies. A central result is the computation of the lean kernel (the largest lean subset) of a (generalised) clause-set in polynomial time for bounded maximal deficiency. Part II Concluding this mini-series of 2 articles on the foundations of generalised clause-sets, we study the combinatorial properties of non-boolean conjunctive normal forms (clause-sets), allowing arbitrary (but finite) sets of values for variables, while literals express that some variable shall not get some (given) value. First we study the properties of the direct translation (or “encoding”) of generalised clausesets into boolean clause-sets. Many combinatorial properties are preserved, and as a result we can lift fixed-parameter tractability of satisfiability in the maximal deficiency from the boolean case to the general case. Then we turn to irredundant clause-sets, which generalise minimally unsatisfiable clause-sets, and we prove basic properties. The simplest irredundant clause-sets are hitting clause-sets, and we provide characterisations and generalisations. Unsatisfiable irredundant clause-sets are the minimally unsatisfiable clause-sets, and we provide basic tools. These tools allow us to characterise the minimally unsatisfiable clause-sets of minimal deficiency. Finally we provide a new translation of generalised boolean clause-sets into boolean clause-sets, the nested translation, which preserves the conflict structure. As an application, we can generalise results for boolean clause-sets regarding the hermitian rank/defect, especially the characterisation of unsatisfiable hitting clause-sets where between every two clauses we have exactly one conflict. We conclude with a list of open problems, and a discussion of the “generic translation scheme”.
منابع مشابه
Constraint - Based Reasoning and Constraint Programming 182
We show the equivalence between the so-called DavisPutnam procedure [3, 2] and the Forward Checking of Haralick and Elliot [7]. Both apply the paradigm choose and propagate in two different formalisms, namely the propositional calculus and the constraint satisfaction problems formalism. They happen to be strictly equivalent as soon as a compatible instantiation order is chosen. This equivalence...
متن کاملFunctional Treewidth: Bounding Complexity in the Presence of Functional Dependencies
Many reasoning problems in logic and constraint satisfaction have been shown to be exponential only in the treewidth of their interaction graph: a graph which captures the structural interactions among variables in a problem. It has long been observed in both logic and constraint satisfaction, however, that problems may be easy even when their treewidth is quite high. To bridge some of the gap ...
متن کاملStructure-Based Local Search Heuristics for Circuit-Level Boolean Satisfiability
This work focuses on improving state-of-the-art in stochastic local search (SLS) for solving Boolean satisfiability (SAT) instances arising from realworld industrial SAT application domains. The recently introduced SLS method CRSAT has been shown to noticeably improve on previously suggested SLS techniques in solving such real-world instances by combining justification-based local search with l...
متن کاملResolution procedures for multiple-valued optimization
Signed clausal forms offer a suitable logical framework for automated reasoning in multiple-valued logics. It turns out that the satisfiability problem of any finitely-valued propositional logic, as well as of certain infinitely-valued logics, can be easily reduced, in polynomial time, to the satisfiability problem of signed clausal forms. On the other hand, signed clausal forms are a powerful ...
متن کاملA Hybrid BDD and SAT Finite Domain Constraint Solver
Finite-domain constraint solvers based on Binary Decision Diagrams (BDDs) are a powerful technique for solving constraint problems over finite set and integer variables represented as Boolean formulæ. Boolean Satisfiability (SAT) solvers are another form of constraint solver that operate on constraints on Boolean variables expressed in clausal form. Modern SAT solvers have highly optimized prop...
متن کاملA New Method for Solving Constraint Satisfaction Problems
Many important problems in Artificial Intelligence can be defined as Constraint Satisfaction Problems (CSP). These types of problems are defined by a limited set of variables, each having a limited domain and a number of Constraints on the values of those variables (these problems are also called Consistent Labeling Problems (CLP), in which “Labeling means assigning a value to a variable.) Solu...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1103.3693 شماره
صفحات -
تاریخ انتشار 2011