Many interesting tractable problems are identified under the model of Constraint Satisfaction Problems. These problems are usually solved by forcing a certain level of local consistency. In this paper, for the class of connected row convex constraints, we propose a novel algorithm which is based on the ideas of variable elimination and efficient composition of row convex and connected constraints. Compared with the existing work including randomized algorithms, the new algorithm has better worst case time and space complexity.
|Number of pages
|IJCAI International Joint Conference on Artificial Intelligence
|Published - 2007
|20th International Joint Conference on Artificial Intelligence, IJCAI 2007 - Hyderabad, India
Duration: Jan 6 2007 → Jan 12 2007