English

A Simple, Nearly-Optimal Algorithm for Differentially Private All-Pairs Shortest Distances

Data Structures and Algorithms 2024-07-16 v2

Abstract

The all-pairs shortest distances (APSD) with differential privacy (DP) problem takes as input an undirected, weighted graph G=(V,E,w)G = (V,E, \mathbf{w}) and outputs a private estimate of the shortest distances in GG between all pairs of vertices. In this paper, we present a simple O~(n1/3/ε)\widetilde{O}(n^{1/3}/\varepsilon)-accurate algorithm to solve APSD with ε\varepsilon-DP, which reduces to O~(n1/4/ε)\widetilde{O}(n^{1/4}/\varepsilon) in the (ε,δ)(\varepsilon, \delta)-DP setting, where n=Vn = |V|. Our algorithm greatly improves upon the error of prior algorithms, namely O~(n2/3/ε)\widetilde{O}(n^{2/3}/\varepsilon) and O~(n/ε)\widetilde{O}(\sqrt{n}/\varepsilon) in the two respective settings, and is the first to be optimal up to a polylogarithmic factor, based on a lower bound of Ω~(n1/4)\widetilde{\Omega}(n^{1/4}). In the case where a multiplicative approximation is allowed, we give two different constructions of algorithms with reduced additive error. Our first construction allows a multiplicative approximation of O(kloglogn)O(k\log{\log{n}}) and has additive error O~(kn1/k/ε)\widetilde{O}(k\cdot n^{1/k}/\varepsilon) in the ε\varepsilon-DP case and O~(kn1/(2k)/ε)\widetilde{O}(\sqrt{k}\cdot n^{1/(2k)}/\varepsilon) in the (ε,δ)(\varepsilon, \delta)-DP case. Our second construction allows multiplicative approximation 2k12k-1 and has the same asymptotic additive error as the first construction. Both constructions significantly improve upon the currently best-known additive error of, O~(kn1/2+1/(4k+2)/ε)\widetilde{O}(k\cdot n^{1/2 + 1/(4k+2)}/\varepsilon) and O~(kn1/3+2/(9k+3)/ε)\widetilde{O}(k\cdot n^{1/3 + 2/(9k+3)}/\varepsilon), respectively. Our algorithms are straightforward and work by decomposing a graph into a set of spanning trees, and applying a key observation that we can privately release APSD in trees with O(polylog(n))O(\text{polylog}(n)) error.

Keywords

Cite

@article{arxiv.2407.06913,
  title  = {A Simple, Nearly-Optimal Algorithm for Differentially Private All-Pairs Shortest Distances},
  author = {Jesse Campbell and Chunjiang Zhu},
  journal= {arXiv preprint arXiv:2407.06913},
  year   = {2024}
}

Comments

Error in Section 3: (1) Improper assumption that the topology of the shortest path trees are public. (2) Improper usage of Lemma 2.4. Error in Section 4: Improper assumption that the topology of the shortest path trees are public

R2 v1 2026-06-28T17:34:26.279Z