English

Improved Lower Bound for Differentially Private Facility Location

Data Structures and Algorithms 2024-03-11 v1 Cryptography and Security

Abstract

We consider the differentially private (DP) facility location problem in the so called super-set output setting proposed by Gupta et al. [SODA 2010]. The current best known expected approximation ratio for an ϵ\epsilon-DP algorithm is O(lognϵ)O\left(\frac{\log n}{\sqrt{\epsilon}}\right) due to Cohen-Addad et al. [AISTATS 2022] where nn denote the size of the metric space, meanwhile the best known lower bound is Ω(1/ϵ)\Omega(1/\sqrt{\epsilon}) [NeurIPS 2019]. In this short note, we give a lower bound of Ω~(min{logn,lognϵ})\tilde{\Omega}\left(\min\left\{\log n, \sqrt{\frac{\log n}{\epsilon}}\right\}\right) on the expected approximation ratio of any ϵ\epsilon-DP algorithm, which is the first evidence that the approximation ratio has to grow with the size of the metric space.

Keywords

Cite

@article{arxiv.2403.04874,
  title  = {Improved Lower Bound for Differentially Private Facility Location},
  author = {Pasin Manurangsi},
  journal= {arXiv preprint arXiv:2403.04874},
  year   = {2024}
}
R2 v1 2026-06-28T15:12:53.910Z