Eigenvector statistics of sparse random matrices

Paul Bourgade, Jiaoyang Huang, Horng Tzer Yau

Research output: Contribution to journalArticle

Abstract

We prove that the bulk eigenvectors of sparse random matrices, i.e. the adjacency matrices of Erdős-Rényi graphs or random regular graphs, are asymptotically jointly normal, provided the averaged degree increases with the size of the graphs. Our methodology follows [6] by analyzing the eigenvector flow under Dyson Brownian motion, combined with an isotropic local law for Green’s function. As an auxiliary result, we prove that for the eigenvector flow of Dyson Brownian motion with general initial data, the eigenvectors are asymptotically jointly normal in the direction q after time η ≪ t ≪ r, if in a window of size r, the initial density of states is bounded below and above down to the scale η, and the initial eigenvectors are delocalized in the direction q down to the scale η.

Original languageEnglish (US)
Article number64
JournalElectronic Journal of Probability
Volume22
DOIs
StatePublished - 2017

Fingerprint

Sparse matrix
Random Matrices
Eigenvector
Statistics
Brownian motion
Adjacency Matrix
Density of States
Graph in graph theory
Regular Graph
Random Graphs
Green's function
Graph
Methodology

Keywords

  • Eigenvectors
  • Isotropic local law
  • Sparse random graphs

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Eigenvector statistics of sparse random matrices. / Bourgade, Paul; Huang, Jiaoyang; Yau, Horng Tzer.

In: Electronic Journal of Probability, Vol. 22, 64, 2017.

Research output: Contribution to journalArticle

@article{a7fa51422fc34465ba43c921eb4a7e3a,
title = "Eigenvector statistics of sparse random matrices",
abstract = "We prove that the bulk eigenvectors of sparse random matrices, i.e. the adjacency matrices of Erdős-R{\'e}nyi graphs or random regular graphs, are asymptotically jointly normal, provided the averaged degree increases with the size of the graphs. Our methodology follows [6] by analyzing the eigenvector flow under Dyson Brownian motion, combined with an isotropic local law for Green’s function. As an auxiliary result, we prove that for the eigenvector flow of Dyson Brownian motion with general initial data, the eigenvectors are asymptotically jointly normal in the direction q after time η∗ ≪ t ≪ r, if in a window of size r, the initial density of states is bounded below and above down to the scale η∗, and the initial eigenvectors are delocalized in the direction q down to the scale η∗.",
keywords = "Eigenvectors, Isotropic local law, Sparse random graphs",
author = "Paul Bourgade and Jiaoyang Huang and Yau, {Horng Tzer}",
year = "2017",
doi = "10.1214/17-EJP81",
language = "English (US)",
volume = "22",
journal = "Electronic Journal of Probability",
issn = "1083-6489",
publisher = "Institute of Mathematical Statistics",

}

TY - JOUR

T1 - Eigenvector statistics of sparse random matrices

AU - Bourgade, Paul

AU - Huang, Jiaoyang

AU - Yau, Horng Tzer

PY - 2017

Y1 - 2017

N2 - We prove that the bulk eigenvectors of sparse random matrices, i.e. the adjacency matrices of Erdős-Rényi graphs or random regular graphs, are asymptotically jointly normal, provided the averaged degree increases with the size of the graphs. Our methodology follows [6] by analyzing the eigenvector flow under Dyson Brownian motion, combined with an isotropic local law for Green’s function. As an auxiliary result, we prove that for the eigenvector flow of Dyson Brownian motion with general initial data, the eigenvectors are asymptotically jointly normal in the direction q after time η∗ ≪ t ≪ r, if in a window of size r, the initial density of states is bounded below and above down to the scale η∗, and the initial eigenvectors are delocalized in the direction q down to the scale η∗.

AB - We prove that the bulk eigenvectors of sparse random matrices, i.e. the adjacency matrices of Erdős-Rényi graphs or random regular graphs, are asymptotically jointly normal, provided the averaged degree increases with the size of the graphs. Our methodology follows [6] by analyzing the eigenvector flow under Dyson Brownian motion, combined with an isotropic local law for Green’s function. As an auxiliary result, we prove that for the eigenvector flow of Dyson Brownian motion with general initial data, the eigenvectors are asymptotically jointly normal in the direction q after time η∗ ≪ t ≪ r, if in a window of size r, the initial density of states is bounded below and above down to the scale η∗, and the initial eigenvectors are delocalized in the direction q down to the scale η∗.

KW - Eigenvectors

KW - Isotropic local law

KW - Sparse random graphs

UR - http://www.scopus.com/inward/record.url?scp=85027300123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027300123&partnerID=8YFLogxK

U2 - 10.1214/17-EJP81

DO - 10.1214/17-EJP81

M3 - Article

VL - 22

JO - Electronic Journal of Probability

JF - Electronic Journal of Probability

SN - 1083-6489

M1 - 64

ER -