Nonreconstruction of high-dimensional stochastic block model with bounded degree

Nonreconstruction of high-dimensional stochastic block model with bounded degree

In this paper, we study the stochastic block model (SBM) with growing number of clusters and bounded degree. Specifically, for SBM Gsn(n,an,bn) with diverging sn blocks and fixed a and b(a>b>0), we prove that if (a−b)2<b, then it is impossible to distinguish Gsn(n,an,bn) from the correspond...

Full description

Saved in:
Bibliographic Details
Published in:Statistics & probability letters Vol. 158; p. 108675
Main Authors: Zhang, Yue, Yuan, Mingao
Format: Journal Article
Language:English
Published: Elsevier B.V 01-03-2020
Subjects:
Community detection
Networks
Random graph
Stochastic block model
secondary
Community detection
Networks
Stochastic block model
primary
Random graph
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we study the stochastic block model (SBM) with growing number of clusters and bounded degree. Specifically, for SBM Gsn(n,an,bn) with diverging sn blocks and fixed a and b(a>b>0), we prove that if (a−b)2<b, then it is impossible to distinguish Gsn(n,an,bn) from the corresponding Erdös–Rényi model Gn,a+(sn−1)bnsn.
ISSN:0167-7152
1879-2103
DOI:10.1016/j.spl.2019.108675