English

Dependable Spanners via Unreliable Edges

Computational Geometry 2025-02-19 v3

Abstract

Let PP be a set of nn points in Rd\mathbb{R}^d, and let ε,ψ(0,1)\varepsilon,\psi \in (0,1) be parameters. Here, we consider the task of constructing a (1+ε)(1+\varepsilon)-spanner for PP, where every edge might fail (independently) with probability 1ψ1-\psi. For example, for ψ=0.1\psi=0.1, about 90%90\% of the edges of the graph fail. Nevertheless, we show how to construct a spanner that survives such a catastrophe with near linear number of edges. The measure of reliability of the graph constructed is how many pairs of vertices lose (1+ε)(1+\varepsilon)-connectivity. Surprisingly, despite the spanner constructed being of near linear size, the number of failed pairs is close to the number of failed pairs if the underlying graph was a clique. Specifically, we show how to construct such an exact dependable spanner in one dimension of size O(nψlogn)O(\tfrac{n}{\psi} \log n), which is optimal. Next, we build an (1+ε)(1+\varepsilon)-spanners for a set PRdP \subseteq \mathbb{R}^d of nn points, of size O(Cnlogn)O( C n \log n ), where C1/(εdψ4/3)C \approx 1/\bigl(\varepsilon^{d} \psi^{4/3}\bigr). Surprisingly, these new spanners also have the property that almost all pairs of vertices have a 4\leq 4-hop paths between them realizing this short path.

Keywords

Cite

@article{arxiv.2407.01466,
  title  = {Dependable Spanners via Unreliable Edges},
  author = {Sariel Har-Peled and Maria C. Lusardi},
  journal= {arXiv preprint arXiv:2407.01466},
  year   = {2025}
}
R2 v1 2026-06-28T17:25:15.268Z