DocumentCode :
2815822
Title :
Combined Strategies for Decomposition-Based Methods for Solving CSPs
Author :
Jegou, Philippe ; Ndiaye, Samba Ndojh ; Terrioux, Cyril
Author_Institution :
LSIS, Univ. Paul Cezanne, Marseille, France
fYear :
2009
fDate :
2-4 Nov. 2009
Firstpage :
184
Lastpage :
192
Abstract :
In this paper, we consider theoretical and practical methods based on decompositions of constraint networks. We exploit the fact that decomposition-based methods can be used considering two steps. The first step is related to the (hyper)graphical decomposition (e.g. tree-decomposition or hyper tree-decomposition) while the second step exploits the decomposition to solve the CSPs. Thanks to this approach, we define then hybrid methods which can be optimal from a theoretical viewpoint while being efficient in practice. The complexity analysis of these combined methods allows us to give a more detailed presentation of the constraint tractability hierarchy introduced in. Finally, we justify our approach with experimental results.
Keywords :
communicating sequential processes; computational complexity; trees (mathematics); communicating sequential processes; complexity analysis; constraint networks; constraint tractability hierarchy; decomposition-based methods; hyper tree-decomposition; Artificial intelligence; Constraint theory; Filtering algorithms; Heuristic algorithms; High definition video; Large scale integration; Magnetohydrodynamics; Optimization methods; Polynomials; Constraint satisfaction; structural methods;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Tools with Artificial Intelligence, 2009. ICTAI '09. 21st International Conference on
Conference_Location :
Newark, NJ
ISSN :
1082-3409
Print_ISBN :
978-1-4244-5619-2
Electronic_ISBN :
1082-3409
Type :
conf
DOI :
10.1109/ICTAI.2009.70
Filename :
5363271
Link To Document :
بازگشت