DMCA
Arc Consistency and Quasigroup Completion (1998)
Cached
Download Links
| Venue: | In Proceedings of the ECAI-98 workshop on non-binary constraints |
| Citations: | 24 - 4 self |
BibTeX
@INPROCEEDINGS{Shaw98arcconsistency,
author = {Paul Shaw and Kostas Stergiou and Toby Walsh},
title = {Arc Consistency and Quasigroup Completion},
booktitle = {In Proceedings of the ECAI-98 workshop on non-binary constraints},
year = {1998}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Quasigroup completion is a recently proposed benchmark constraint satisfaction problem that combines the features of randomly generated instances and highly structured problems. A quasigroup completion problem can be represented as a CSP with n 2 variables, each with a domain of size n. The constraints can be represented either by 2n all-different n-ary constraints or by binary pairwise constraints, giving a constraint graph with 2n cliques of size n. We present a comparison between the two representations and show that the n\Gammaary representation reduces the cost of solving quasigroup completion problems drastically. In addition, our study shows that while it is desirable to add structure to random problems, we must be careful that the added structure does not make the problems trivial, given the right constraint propagators. Introduction Quasigroup completion (GS97b; GS97a; GSC97) is a recently proposed benchmark constraint satisfaction problem that combines the features of ra...
Keyphrases
quasigroup completion arc consistency benchmark constraint satisfaction problem quasigroup completion problem right constraint propagator structured problem added structure all-different n-ary constraint introduction quasigroup completion gammaary representation binary pairwise constraint constraint graph
