Facility Location Problem in Differential Privacy Model Revisited

Authors: Yunus Esencayi, Marco Gaboardi, Shi Li, Di Wang

NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper we study the uncapacitated facility location problem in the model of differential privacy (DP) with uniform facility cost. Specifically, we first show that, under the hierarchically well-separated tree (HST) metrics and the super-set output setting that was introduced in [8], there is an ϵ-DP algorithm that achieves an O( 1ϵ ) (expected multiplicative) approximation ratio; this implies an O( log nϵ ) approximation ratio for the general metric case, where n is the size of the input metric. These bounds improve the best-known results given by [8].
Researcher Affiliation Academia Yunus Esencayi SUNY at Buffalo yunusese@buffalo.edu Marco Gaboardi Boston University gaboardi@bu.edu Shi Li SUNY at Buffalo shil@buffalo.edu Di Wang SUNY at Buffalo dwang45@buffalo.edu
Pseudocode Yes Algorithm 1 UFL-tree-base(ϵ) and Algorithm 2 DP-UFL-tree(ϵ)
Open Source Code No The paper does not provide any concrete access information (e.g., a specific link or explicit statement of code release) for open-source code.
Open Datasets No This is a theoretical paper focused on algorithms and proofs, not empirical evaluation on datasets. Therefore, it does not mention training data availability.
Dataset Splits No This is a theoretical paper focused on algorithms and proofs, not empirical evaluation on datasets. Therefore, it does not mention validation splits.
Hardware Specification No This is a theoretical paper and does not describe experiments that would require hardware specifications.
Software Dependencies No This is a theoretical paper and does not mention specific software dependencies with version numbers.
Experiment Setup No This is a theoretical paper and does not describe experimental setups with hyperparameters or training configurations.