Fixed-parameter tractability of satisfying beyond the number of variables
From MaRDI portal
Publication:528862
DOI10.1007/S00453-012-9697-4zbMATH Open1360.68502arXiv1212.0106OpenAlexW3099433168MaRDI QIDQ528862FDOQ528862
Publication date: 17 May 2017
Published in: Algorithmica (Search for Journal in Brave)
Abstract: We consider a CNF formula as a multiset of clauses: . The set of variables of will be denoted by . Let denote the bipartite graph with partite sets and and with an edge between and if or . The matching number of is the size of a maximum matching in . In our main result, we prove that the following parameterization of {sc MaxSat} (denoted by - extsc{SAT}) is fixed-parameter tractable: Given a formula , decide whether we can satisfy at least clauses in , where is the parameter. A formula is called variable-matched if Let and Our main result implies fixed-parameter tractability of {sc MaxSat} parameterized by for variable-matched formulas ; this complements related results of Kullmann (2000) and Szeider (2004) for {sc MaxSat} parameterized by . To obtain our main result, we reduce - extsc{SAT} into the following parameterization of the {sc Hitting Set} problem (denoted by -{sc Hitting Set}): given a collection of subsets of a ground set of elements, decide whether there is such that for each and where is the parameter. Gutin, Jones and Yeo (2011) proved that -{sc Hitting Set} is fixed-parameter tractable by obtaining an exponential kernel for the problem. We obtain two algorithms for -{sc Hitting Set}: a deterministic algorithm of runtime and a randomized algorithm of expected runtime . Our deterministic algorithm improves an algorithm that follows from the kernelization result of Gutin, Jones and Yeo (2011).
Full work available at URL: https://arxiv.org/abs/1212.0106
Cites Work
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- The complexity of facets resolved
- Parameterizing above or below guaranteed values
- On problems without polynomial kernels
- A new lower bound on the maximum number of satisfied clauses in Max-SAT and its algorithmic applications
- Parametrized complexity theory.
- Incompressibility through Colors and IDs
- Parameterizing above Guaranteed Values: MaxSat and MaxCut
- Color-coding
- Solving MAX-\(r\)-SAT above a tight lower bound
- Solving satisfiability in less than \(2^ n\) steps
- Faster algorithms for finding and counting subgraphs
- Kernel bounds for disjoint cycles and disjoint paths
- Minimal non-two-colorable hypergraphs and minimal unsatisfiable formulas
- Lean clause-sets: Generalizations of minimally unsatisfiable clause-sets
- Minimal unsatisfiable formulas with bounded clause-variable difference are fixed-parameter tractable
- On subclasses of minimal unsatisfiable formulas
- Polynomial-time recognition of minimal unsatisfiable formulas with fixed clause-variable difference.
- A New Bound for 3-Satisfiable Maxsat and Its Algorithmic Application
- Kernels for below-upper-bound parameterizations of the hitting set and directed dominating set problems
Cited In (2)
This page was built for publication: Fixed-parameter tractability of satisfying beyond the number of variables
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q528862)