Non uniform selection of solutions for upper bounding the 3-SAT threshold

Authors:
Yacine Boufkhad;Thomas Hugel
Affiliations:
LIAFA, Université Denis Diderot Paris 7 - CNRS, Paris;LIAFA, Université Denis Diderot Paris 7 - CNRS, Paris
Venue:
SAT'10 Proceedings of the 13th international conference on Theory and Applications of Satisfiability Testing
Year:
2010

Citing 7
Cited 1

Tail bounds for occupancy and the satisfiability threshold conjecture

Random Structures & Algorithms
A general upper bound for the satisfiability threshold of random r-SAT formulae

Journal of Algorithms
Approximating the unsatisfiability threshold of random formulas

Random Structures & Algorithms
Typical random 3-SAT formulae and the satisfiability threshold

SODA '00 Proceedings of the eleventh annual ACM-SIAM symposium on Discrete algorithms
Regular Random k-SAT: Properties of Balanced Formulas

Journal of Automated Reasoning
On the satisfiability threshold of formulas with three literals per clause

Theoretical Computer Science
Hard and easy distributions of SAT problems

AAAI'92 Proceedings of the tenth national conference on Artificial intelligence

Estimating satisfiability

Discrete Applied Mathematics

Quantified Score

Hi-index	0.00

Visualization

Abstract

We give a new insight into the upper bounding of the 3-SAT threshold by the first moment method. The best criteria developed so far to select the solutions to be counted discriminate among neighboring solutions on the basis of uniform information about each individual free variable. What we mean by uniform information, is information which does not depend on the solution: e.g. the number of positive/negative occurrences of the considered variable. What is new in our approach is that we use non uniform information about variables. Thus we are able to make a more precise tuning, resulting in a slight improvement on upper bounding the 3-SAT threshold for various models of formulas defined by their distributions.