English

A Schur Complement Cheeger Inequality

Discrete Mathematics 2018-11-28 v1 Data Structures and Algorithms Probability

Abstract

Cheeger's inequality shows that any undirected graph GG with minimum nonzero normalized Laplacian eigenvalue λG\lambda_G has a cut with conductance at most O(λG)O(\sqrt{\lambda_G}). Qualitatively, Cheeger's inequality says that if the relaxation time of a graph is high, there is a cut that certifies this. However, there is a gap in this relationship, as cuts can have conductance as low as Θ(λG)\Theta(\lambda_G). To better approximate the relaxation time of a graph, we consider a more general object. Instead of bounding the mixing time with cuts, we bound it with cuts in graphs obtained by Schur complementing out vertices from the graph GG. Combinatorially, these Schur complements describe random walks in GG restricted to a subset of its vertices. As a result, all Schur complement cuts have conductance at least Ω(λG)\Omega(\lambda_G). We show that unlike with cuts, this inequality is tight up to a constant factor. Specifically, there is a Schur complement cut with conductance at most O(λG)O(\lambda_G).

Keywords

Cite

@article{arxiv.1811.10834,
  title  = {A Schur Complement Cheeger Inequality},
  author = {Aaron Schild},
  journal= {arXiv preprint arXiv:1811.10834},
  year   = {2018}
}

Comments

ITCS 2019

R2 v1 2026-06-23T06:21:34.566Z