English

Dynamic Algorithms for Matroid Submodular Maximization

Data Structures and Algorithms 2023-12-27 v2 Machine Learning

Abstract

Submodular maximization under matroid and cardinality constraints are classical problems with a wide range of applications in machine learning, auction theory, and combinatorial optimization. In this paper, we consider these problems in the dynamic setting, where (1) we have oracle access to a monotone submodular function f:2VR+f: 2^{V} \rightarrow \mathbb{R}^+ and (2) we are given a sequence S\mathcal{S} of insertions and deletions of elements of an underlying ground set VV. We develop the first fully dynamic (4+ϵ)(4+\epsilon)-approximation algorithm for the submodular maximization problem under the matroid constraint using an expected worst-case O(klog(k)log3(k/ϵ))O(k\log(k)\log^3{(k/\epsilon)}) query complexity where 0<ϵ10 < \epsilon \le 1. This resolves an open problem of Chen and Peng (STOC'22) and Lattanzi et al. (NeurIPS'20). As a byproduct, for the submodular maximization under the cardinality constraint kk, we propose a parameterized (by the cardinality constraint kk) dynamic algorithm that maintains a (2+ϵ)(2+\epsilon)-approximate solution of the sequence S\mathcal{S} at any time tt using an expected worst-case query complexity O(kϵ1log2(k))O(k\epsilon^{-1}\log^2(k)). This is the first dynamic algorithm for the problem that has a query complexity independent of the size of ground set VV.

Keywords

Cite

@article{arxiv.2306.00959,
  title  = {Dynamic Algorithms for Matroid Submodular Maximization},
  author = {Kiarash Banihashem and Leyla Biabani and Samira Goudarzi and MohammadTaghi Hajiaghayi and Peyman Jabbarzade and Morteza Monemizadeh},
  journal= {arXiv preprint arXiv:2306.00959},
  year   = {2023}
}