English

Improved Lower Bounds for Submodular Function Minimization

Data Structures and Algorithms 2022-07-12 v1 Computational Complexity Distributed, Parallel, and Cluster Computing Discrete Mathematics Optimization and Control

Abstract

We provide a generic technique for constructing families of submodular functions to obtain lower bounds for submodular function minimization (SFM). Applying this technique, we prove that any deterministic SFM algorithm on a ground set of nn elements requires at least Ω(nlogn)\Omega(n \log n) queries to an evaluation oracle. This is the first super-linear query complexity lower bound for SFM and improves upon the previous best lower bound of 2n2n given by [Graur et al., ITCS 2020]. Using our construction, we also prove that any (possibly randomized) parallel SFM algorithm, which can make up to poly(n)\mathsf{poly}(n) queries per round, requires at least Ω(n/logn)\Omega(n / \log n) rounds to minimize a submodular function. This improves upon the previous best lower bound of Ω~(n1/3)\tilde{\Omega}(n^{1/3}) rounds due to [Chakrabarty et al., FOCS 2021], and settles the parallel complexity of query-efficient SFM up to logarithmic factors due to a recent advance in [Jiang, SODA 2021].

Keywords

Cite

@article{arxiv.2207.04342,
  title  = {Improved Lower Bounds for Submodular Function Minimization},
  author = {Deeparnab Chakrabarty and Andrei Graur and Haotian Jiang and Aaron Sidford},
  journal= {arXiv preprint arXiv:2207.04342},
  year   = {2022}
}

Comments

To appear in FOCS 2022