• DocumentCode
    2261938
  • Title

    Approximation algorithms for MAX SAT: Yannakakis vs. Goemans-Williamson

  • Author

    Asano, Takao

  • Author_Institution
    Dept. of Inf. & Syst. Eng., Chuo Univ., Tokyo, Japan
  • fYear
    1997
  • fDate
    17-19 Jun 1997
  • Firstpage
    24
  • Lastpage
    37
  • Abstract
    MAX SAT (the maximum satisfiability problem) is stated as follows: given a set of clauses with weights, find a truth assignment that maximizes the sum of the weights of the satisfied clauses. In this paper, we consider approximation algorithms for MAX SAT proposed by Yannnkakis and Goemans-Williamson and present an approximation algorithm which is an improvement of Yannakakis´ algorithm. Although Yannakakis´ original algorithm has no better performance guarantee than Goemans-Williamson, our improved algorithm has a better performance guarantee and leads to a 0.770 approximation algorithm
  • Keywords
    approximation theory; computability; computational complexity; MAX SAT; approximation algorithms; clauses; maximum satisfiability problem; truth assignment; weights; Approximation algorithms; Educational programs; Polynomials; Programming profession; Systems engineering and theory;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Theory of Computing and Systems, 1997., Proceedings of the Fifth Israeli Symposium on
  • Conference_Location
    Ramat-Gan
  • Print_ISBN
    0-8186-8037-7
  • Type

    conf

  • DOI
    10.1109/ISTCS.1997.595154
  • Filename
    595154