Up and Down Mini-Buckets: A Scheme for Approximating Combinatorial Optimization Tasks (2001)
| Citations: | 10 - 4 self |
BibTeX
@TECHREPORT{Kask01upand,
author = {Kalev Kask and Javier Larrosa and Rina Dechter},
title = {Up and Down Mini-Buckets: A Scheme for Approximating Combinatorial Optimization Tasks},
institution = {},
year = {2001}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
The paper addresses the problem of computing lower bounds on the optimal costs associated with each unary assignment of a value to a variable in combinatorial optimization problems. This task is instrumental in probabilistic reasoning and is also important for the development of admissible heuristic functions that can guide search algorithms for optimal solutions. The paper presents UD-MB, a new algorithm that applies the mini-bucket elimination idea [Dechter and Rish, 1997] to accomplish this task. We show empirically that UD-MB may achieve a substantial speed up over a bruteforce approximation method via mini-buckets. 1 Introduction A Combinatorial optimization problem (COP) is defined by a finite set of variables, a set of finite domains and a set of cost functions. A solution is the optimal cost over the set of complete assignments of values to variables. It is well known that solving COP is NP-hard. COP arise in a large variety of domains including artificial intelli...
Citations
No citations identified.
