Skip navigation
Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp018623j209r
Title: Estimating Block Model Graphons via Spectral Recovery in the Irregular Sparse Setting
Authors: Kelley, Anna
Advisors: Sly, Allan
Department: Mathematics
Certificate Program: Center for Statistics and Machine Learning
Class Year: 2024
Abstract: The problem of estimating graphons extends the popular topic of Stochastic Block Model (SBM) recovery to the continuous setting. Spectral recovery methods are particularly useful as they are direct and generalizable to the graphon setting. Past sparse SBM and graphon research developed spectral recovery algorithms assuming regular expected degree. Meanwhile, algorithms for SBM community recovery in the irregular setting rely on more involved techniques exploiting the block structure. We present a spectral algorithm for asymptotic estimation of sparse block model graphons and recovery of SBM communities with no assumption of regularity. We discuss the implications of this finding for the estimation of general sparse irregular graphons.
URI: http://arks.princeton.edu/ark:/88435/dsp018623j209r
Type of Material: Princeton University Senior Theses
Language: en
Appears in Collections:Mathematics, 1934-2024

Files in This Item:
File Description SizeFormat 
KELLEY-ANNA-THESIS.pdf540.65 kBAdobe PDF    Request a copy


Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.