4.4 Article

Submodular Maximization over Multiple Matroids via Generalized Exchange Properties

Journal

MATHEMATICS OF OPERATIONS RESEARCH
Volume 35, Issue 4, Pages 795-806

Publisher

INFORMS
DOI: 10.1287/moor.1100.0463

Keywords

matroid; submodular function; approximation algorithm

Ask authors/readers for more resources

Submodular function maximization is a central problem in combinatorial optimization, generalizing many important NP-hard problems including max cut in digraphs, graphs, and hypergraphs; certain constraint satisfaction problems; maximum entropy sampling; and maximum facility location problems. Our main result is that for any k >= 2 and any epsilon > 0, there is a natural local search algorithm that has approximation guarantee of 1/(k+epsilon) for the problem of maximizing a monotone submodular function subject to k matroid constraints. This improves upon the 1/(k+1)-approximation of Fisher, Nemhauser, and Wolsey obtained in 1978 [Fisher, M., G. Nemhauser, L. Wolsey. 1978. An analysis of approximations for maximizing submodular set functions-II. Math. Programming Stud. 8 73-87]. Also, our analysis can be applied to the problem of maximizing a linear objective function and even a general nonmonotone submodular function subject to k matroid constraints. We show that, in these cases, the approximation guarantees of our algorithms are 1/(k-1+epsilon) and 1/(k+1+1/(k-1)+epsilon), respectively. Our analyses are based on two new exchange properties for matroids. One is a generalization of the classical Rota exchange property for matroid bases, and another is an exchange property for two matroids based on the structure of matroid intersection.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.4
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available