Explaining flow-based propagation

Authors:
Nicholas Downing;Thibaut Feydy;Peter J. Stuckey
Affiliations:
National ICT Australia and The University of Melbourne, Victoria, Australia;National ICT Australia and The University of Melbourne, Victoria, Australia;National ICT Australia and The University of Melbourne, Victoria, Australia
Venue:
CPAIOR'12 Proceedings of the 9th international conference on Integration of AI and OR Techniques in Constraint Programming for Combinatorial Optimization Problems
Year:
2012

Citing 11
Cited 0

Chaff: engineering an efficient SAT solver

Proceedings of the 38th annual Design Automation Conference
CSPLIB: A Benchmark Library for Constraints

CP '99 Proceedings of the 5th International Conference on Principles and Practice of Constraint Programming
Network Flow Problems in Constraint Programming

CP '01 Proceedings of the 7th International Conference on Principles and Practice of Constraint Programming
Efficient Intelligent Backtracking Using Linear Programming

INFORMS Journal on Computing
Flow-Based Propagators for the SEQUENCE and Related Global Constraints

CP '08 Proceedings of the 14th international conference on Principles and Practice of Constraint Programming
Propagation via lazy clause generation

Constraints
Encodings of the SEQUENCE constraint

CP'07 Proceedings of the 13th international conference on Principles and practice of constraint programming
Generalized arc consistency for global cardinality constraint

AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1
Nogood processing in csps

Nogood processing in csps
An efficient generic network flow constraint

Proceedings of the 2011 ACM Symposium on Applied Computing
Conflict analysis in mixed integer programming

Discrete Optimization

Quantified Score

Hi-index	0.00

Visualization

Abstract

Lazy clause generation is a powerful approach to reducing search in constraint programming. For use in a lazy clause generation solver, global constraints must be extended to explain themselves. In this paper we present two new generic flow-based propagators (for hard and soft flow-based constraints) with several novel features, and most importantly, the addition of explanation capability. We discuss how explanations change the tradeoffs for propagation compared with the previous generic flow-based propagator, and show that the generic propagators can efficiently replace specialized versions, in particular for gcc and sequence constraints. Using real-world scheduling and rostering problems as examples, we compare against a number of standard Constraint Programming implementations of these contraints (and in the case of soft constraints, Mixed-Integer Programming models) to show that the new global propagators are extremely beneficial on these benchmarks.