Sampling Techniques for Constraint Satisfaction and Beyond

Опубликовано 21 июня 2016, 23:57

Constraint problems have played a key role in diverse areas spanning testing, formal verification, planning, inferencing and the like. Apart from the classical problem of checking satisfiability, the problems of generating satisfying assignments randomly and of counting the total number of satisfying assignments have also attracted significant theoretical and practical interest over the years. Prior work offered heuristic approaches with no guarantee of solution distribution, and approaches with proven guarantees, but poor performance in practice. In this talk, I will describe a novel approach based on limited independent hashing and present two practical algorithms, UniGen and WeightMC, for solving these two fundamental problems. Unlike prior work, our algorithms provide strong theoretical guarantees and also scale to large problem sizes.