An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP

This paper deals with methods exploiting tree-decomposition approaches for solving constraint networks. We consider here the practical efficiency of these approaches by defining five classes of variable orders more and more dynamic which guarantee time complexity bounds from O(exp(w + 1)) to O(exp(2(w + k))), with w the "tree-width" of a CSP and k a constant. Finally, we assess practically their relevance.

• Type: Conference
• Year: 2006
• Where: CP
• Authors: Philippe Jégou, Samba Ndiaye, Cyril Terrioux
• Keywords: Artificial Intelligence, Constraint Networks, CP 2006, Methods Exploiting Tree-decomposition, Time Complexity Bounds