English

Subquadratic Submodular Maximization with a General Matroid Constraint

Data Structures and Algorithms 2024-05-02 v1

Abstract

We consider fast algorithms for monotone submodular maximization with a general matroid constraint. We present a randomized (11/eϵ)(1 - 1/e - \epsilon)-approximation algorithm that requires O~ϵ(rn)\tilde{O}_{\epsilon}(\sqrt{r} n) independence oracle and value oracle queries, where nn is the number of elements in the matroid and rnr \leq n is the rank of the matroid. This improves upon the previously best algorithm by Buchbinder-Feldman-Schwartz [Mathematics of Operations Research 2017] that requires O~ϵ(r2+rn)\tilde{O}_{\epsilon}(r^2 + \sqrt{r}n) queries. Our algorithm is based on continuous relaxation, as with other submodular maximization algorithms in the literature. To achieve subquadratic query complexity, we develop a new rounding algorithm, which is our main technical contribution. The rounding algorithm takes as input a point represented as a convex combination of tt bases of a matroid and rounds it to an integral solution. Our rounding algorithm requires O~(r3/2t)\tilde{O}(r^{3/2} t) independence oracle queries, while the previously best rounding algorithm by Chekuri-Vondr\'{a}k-Zenklusen [FOCS 2010] requires O(r2t)O(r^2 t) independence oracle queries. A key idea in our rounding algorithm is to use a directed cycle of arbitrary length in an auxiliary graph, while the algorithm of Chekuri-Vondr\'{a}k-Zenklusen focused on directed cycles of length two.

Keywords

Cite

@article{arxiv.2405.00359,
  title  = {Subquadratic Submodular Maximization with a General Matroid Constraint},
  author = {Yusuke Kobayashi and Tatsuya Terao},
  journal= {arXiv preprint arXiv:2405.00359},
  year   = {2024}
}

Comments

19 pages, To appear in ICALP 2024

R2 v1 2026-06-28T16:12:31.413Z