English

Maximal $k$-Edge-Connected Subgraphs in Almost-Linear Time for Small $k$

Data Structures and Algorithms 2023-07-04 v1

Abstract

We give the first almost-linear time algorithm for computing the \emph{maximal kk-edge-connected subgraphs} of an undirected unweighted graph for any constant kk. More specifically, given an nn-vertex mm-edge graph G=(V,E)G=(V,E) and a number k=logo(1)nk = \log^{o(1)}n, we can deterministically compute in O(m+n1+o(1))O(m+n^{1+o(1)}) time the unique vertex partition {V1,,Vz}\{V_{1},\dots,V_{z}\} such that, for every ii, ViV_{i} induces a kk-edge-connected subgraph while every superset ViViV'_{i}\supset V_{i} does not. Previous algorithms with linear time work only when k2k\le2 {[}Tarjan SICOMP'72{]}, otherwise they all require Ω(m+nn)\Omega(m+n\sqrt{n}) time even when k=3k=3 {[}Chechik~et~al.~SODA'17; Forster~et~al.~SODA'20{]}. Our algorithm also extends to the decremental graph setting; we can deterministically maintain the maximal kk-edge-connected subgraphs of a graph undergoing edge deletions in m1+o(1)m^{1+o(1)} total update time. Our key idea is a reduction to the dynamic algorithm supporting pairwise kk-edge-connectivity queries {[}Jin and Sun FOCS'20{]}.

Keywords

Cite

@article{arxiv.2307.00147,
  title  = {Maximal $k$-Edge-Connected Subgraphs in Almost-Linear Time for Small $k$},
  author = {Thatchaphol Saranurak and Wuwei Yuan},
  journal= {arXiv preprint arXiv:2307.00147},
  year   = {2023}
}

Comments

Accepted to ESA 2023

R2 v1 2026-06-28T11:19:27.190Z