Chaff: engineering an efficient SAT solver
Proceedings of the 38th annual Design Automation Conference
Propagation via lazy clause generation
Constraints
Lazy clause generation reengineered
CP'09 Proceedings of the 15th international conference on Principles and practice of constraint programming
Why cumulative decomposition is not as bad as it sounds
CP'09 Proceedings of the 15th international conference on Principles and practice of constraint programming
Fast set bounds propagation using a BDD-SAT hybrid
Journal of Artificial Intelligence Research
| Hi-index | 0.00 |
Finite domain propagation solving, the basis of constraint programming (CP) solvers, allows building very high-level models of problems, and using highly specific inference encapsulated in complex global constraints, as well as programming the search for solutions to take into account problem structure. Boolean satisfiability (SAT) solving allows the construction of a graph of inferences made in order to determine and record effective nogoods which prevent the searching of similar parts of the problem, as well as the determination of those variables which form a tightly connected hard part of the problem, thus allowing highly effective automatic search strategies concentrating on these hard parts. Lazy clause generation is a hybrid of CP and SAT solving that combines the strengths of the two approaches. It provides state-of-the-art solutions for a number of hard combinatorial optimization and satisfaction problems. In this invited talk we explain lazy clause generation, and explore some of the many design choices in building such a hybrid system, we also discuss how to further incorporate mixed integer programming (MIP) solving to see if we can also inherit its advantages in combinatorial optimization.