On Identifying Simple and Quantified Lattice Points in the 2SAT Polytope

  • Authors:
  • K. Subramani

  • Affiliations:
  • -

  • Venue:
  • AISC '02/Calculemus '02 Proceedings of the Joint International Conferences on Artificial Intelligence, Automated Reasoning, and Symbolic Computation
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper is concerned with techniques for identifying simple and quantified lattice points in 2SAT polytopes. 2SAT polytopes generalize the polyhedra corresponding to Boolean 2SAT formulae, Vertex-Packing (Covering, Partitioning) and Network flow problems; they find wide application in the domains of Program verification (Software Engineering) and State-Space search (Artificial Intelligence). Our techniques are based on the symbolic elimination strategy called the Fourier-Motzkin elimination procedure and thus have the advantages of being extremely simple (from an implementational perspective) and incremental. We also provide a characterization of the 2SAT polytope in terms of its extreme points and derive some interesting hardness results for associated optimization problems.