, Volume 64, Issue 1, pp 56–68 | Cite as

A New Lower Bound on the Maximum Number of Satisfied Clauses in Max-SAT and Its Algorithmic Applications

  • Robert Crowston
  • Gregory Gutin
  • Mark Jones
  • Anders Yeo


A pair of unit clauses is called conflicting if it is of the form (x), \((\bar{x})\). A CNF formula is unit-conflict free (UCF) if it contains no pair of conflicting unit clauses. Lieberherr and Specker (J. ACM 28:411–421, 1981) showed that for each UCF CNF formula with m clauses we can simultaneously satisfy at least \(\hat{ \varphi } m\) clauses, where \(\hat{ \varphi }=(\sqrt{5}-1)/2\). We improve the Lieberherr-Specker bound by showing that for each UCF CNF formula F with m clauses we can find, in polynomial time, a subformula F′ with m′ clauses such that we can simultaneously satisfy at least \(\hat{ \varphi } m+(1-\hat{ \varphi })m'+(2-3\hat {\varphi })n''/2\) clauses (in F), where n″ is the number of variables in F which are not in F′.

We consider two parameterized versions of MAX-SAT, where the parameter is the number of satisfied clauses above the bounds m/2 and \(m(\sqrt{5}-1)/2\). The former bound is tight for general formulas, and the later is tight for UCF formulas. Mahajan and Raman (J. Algorithms 31:335–354, 1999) showed that every instance of the first parameterized problem can be transformed, in polynomial time, into an equivalent one with at most 6k+3 variables and 10k clauses. We improve this to 4k variables and \((2\sqrt{5}+4)k\) clauses. Mahajan and Raman conjectured that the second parameterized problem is fixed-parameter tractable (FPT). We show that the problem is indeed FPT by describing a polynomial-time algorithm that transforms any problem instance into an equivalent one with at most \((7+3\sqrt{5})k\) variables. Our results are obtained using our improvement of the Lieberherr-Specker bound above.


MaxSat Lower bound 2-Satisfiable Fixed parameter tractable Kernel 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Abu-Khzam, F.N., Fernau, H.: Kernels: annotated, proper and induced. In: Proc. IWPEC 2006. Lect. Notes Comput. Sci., vol. 4169, pp. 264–275 (2006) Google Scholar
  2. 2.
    Aharoni, R., Linial, N.: Minimal non-two-colorable hypergraphs and minimal unsatisfiable formulas. J. Comb. Theory, Ser. A 43(2), 196–204 (1986) MathSciNetzbMATHCrossRefGoogle Scholar
  3. 3.
    Chen, Y., Flum, J., Müller, M.: Lower bounds for kernelizations and other preprocessing procedures. In: Proc. CiE 2009. Lect. Notes Comput. Sci., vol. 5635. pp. 118–128 (2009) Google Scholar
  4. 4.
    Downey, R.G., Fellows, M.R.: Parameterized Complexity. Springer, Berlin (1999) CrossRefGoogle Scholar
  5. 5.
    Flum, J., Grohe, M.: Parameterized Complexity Theory. Springer, Berlin (2006) Google Scholar
  6. 6.
    Fleischner, H., Kullmann, O., Szeider, S.: Polynomial-time recognition of minimal unsatisfiable formulas with fixed clause-variable difference. Theor. Comput. Sci. 289(1), 503–516 (2002) MathSciNetzbMATHCrossRefGoogle Scholar
  7. 7.
    Gutin, G., Jones, M., Yeo, A.: A new bound for 3-satisfiable MaxSat and its algorithmic application. In: Proc. FCT 2011. Lect. Notes Comput. Sci., vol. 6914, pp. 138–147 (2011) Google Scholar
  8. 8.
    Huang, M.A., Lieberherr, K.J.: Implications of forbidden structures for extremal algorithmic problems. Theor. Comput. Sci. 40, 195–210 (1985) MathSciNetzbMATHCrossRefGoogle Scholar
  9. 9.
    Käppeli, C., Scheder, D.: Partial satisfaction of k-satisfiable formulas. Electron. Notes Discrete Math. 29, 497–501 (2007) CrossRefGoogle Scholar
  10. 10.
    Král, D.: Locally satisfiable formulas. In: Proc. SODA 2004, pp. 330–339 (2004) Google Scholar
  11. 11.
    Lieberherr, K.J., Specker, E.: Complexity of partial satisfaction. J. ACM 28(2), 411–421 (1981) MathSciNetzbMATHCrossRefGoogle Scholar
  12. 12.
    Lieberherr, K.J., Specker, E.: Complexity of partial satisfaction, II. Tech. Report 293, Dept. of EECS, Princeton Univ. (1982) Google Scholar
  13. 13.
    Lokshtanov, D.: New methods in parameterized algorithms and complexity. PhD thesis, Bergen (April 2009) Google Scholar
  14. 14.
    Mahajan, M., Raman, V.: Parameterizing above guaranteed values: MaxSat and MaxCut. J. Algorithms 31(2), 335–354 (1999) MathSciNetzbMATHCrossRefGoogle Scholar
  15. 15.
    Monien, B., Speckenmeyer, E.: Solving satisfiability in less than 2n steps. Discrete Appl. Math. 10, 287–295 (1985) MathSciNetzbMATHCrossRefGoogle Scholar
  16. 16.
    Niedermeier, R.: Invitation to Fixed-Parameter Algorithms. Oxford University Press, London (2006) zbMATHCrossRefGoogle Scholar
  17. 17.
    Szeider, S.: Minimal unsatisfiable formulas with bounded clause-variable difference are fixed-parameter tractable. J. Comput. Syst. Sci. 69(4), 656–674 (2004) MathSciNetzbMATHCrossRefGoogle Scholar
  18. 18.
    Trevisan, L.: On local versus global satisfiability. SIAM J. Discrete Math. 17(4), 541–547 (2004) MathSciNetzbMATHCrossRefGoogle Scholar
  19. 19.
    Yannakakis, M.: On the approximation of maximum satisfiability. J. Algorithms 17, 475–502 (1994) MathSciNetzbMATHCrossRefGoogle Scholar
  20. 20.
    West, D.B.: Introduction to Graph Theory, 2nd edn. Prentice Hall, New York (2001) Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Robert Crowston
    • 1
  • Gregory Gutin
    • 1
  • Mark Jones
    • 1
  • Anders Yeo
    • 1
  1. 1.Department of Computer ScienceRoyal Holloway, University of LondonEghamUK

Personalised recommendations