English

Faster Multi-Source Reachability and Approximate Distances via Shortcuts, Hopsets and Matrix Multiplication

Data Structures and Algorithms 2026-05-12 v3 Distributed, Parallel, and Cluster Computing

Abstract

Given an nn-vertex mm-edge digraph G=(V,E)G = (V,E) and a subset SVS \subseteq V of S=nσ|S| = n^{\sigma} (for some 0σ10 \le \sigma \le 1) designated sources, the S×VS \times V reachability problem is to compute the sets Vs\mathcal V_s of vertices reachable from ss, for every sSs \in S. Naive centralized algorithms run BFS/DFS from each source in O(mnσ)O(m \cdot n^{\sigma}) time or compute GG's transitive closure in O^(nω)\hat O(n^{\omega}) time, where ω2.371552\omega \le 2.371552\ldots is the matrix multiplication exponent. Thus, the best known bound is O^(nmin{2+σ,ω})\hat O(n^{\min \{ 2 + \sigma, \omega\}}). Leveraging shortcut constructions by Kogan and Parter [SODA 2022, ICALP 2022], we develop a centralized algorithm with running time O^(n1+23ω(σ))\hat O(n^{1 + \frac{2}{3} \omega(\sigma)}), where ω(σ)\omega(\sigma) is the rectangular matrix multiplication exponent. Using current estimates on ω(σ)\omega(\sigma), our exponent improves upon min{2+σ,ω}\min \{2 + \sigma, \omega \} for σ~σ0.53\tilde \sigma \leq \sigma \leq 0.53, where 1/3<σ~<0.33361/3 < \tilde \sigma < 0.3336 is a universal constant. In a classical result, Cohen [Journal of Algorithms, 1996] devised parallel algorithms for S×VS \times V reachability on graphs admitting balanced recursive separators of size nρn^{\rho} for ρ<1\rho < 1, requiring polylogarithmic time and work nmax{ωρ,2ρ+σ}+o(1)n^{\max \{\omega \rho, 2\rho + \sigma \} + o(1)}. We significantly improve, extend, and generalize Cohen's result. First, our parallel algorithm for graphs with small recursive separators has lower work complexity than Cohen's in boraod paramater ranges. Second, we generalize our algorithm to graphs of treewidth at most nρn^{\rho} (ρ<1\rho < 1) and provide a centralized algorithm that outperforms existing bounds for S×VS \times V reachability on such graphs. We also do this for some other graph familes with small separators. Finally, we extend these results to (1+ϵ)(1 + \epsilon)-approximate distance computation.

Keywords

Cite

@article{arxiv.2507.13470,
  title  = {Faster Multi-Source Reachability and Approximate Distances via Shortcuts, Hopsets and Matrix Multiplication},
  author = {Michael Elkin and Chhaya Trehan},
  journal= {arXiv preprint arXiv:2507.13470},
  year   = {2026}
}
R2 v1 2026-07-01T04:06:52.540Z