Discrete duality principle in different random graph models

For a given random graph, a connected component that contains a finite fraction of the entire graph's vertices is called giant. The study of these components started with the \ER model, where it has been proven that removing the (unique) giant component from a random graph is essentially equiva...

ver descrição completa

Detalhes bibliográficos
Autor: Lesgourgues, Thomas
Formato: tesis de maestría
Fecha de publicación:2019
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/166025
Acesso em linha:https://hdl.handle.net/2117/166025
Access Level:acceso abierto
Palavra-chave:Graph theory
Random graph
Giant Component
Discrete Duality Principle
Branching process
\ER model
Configuration model
Switching
Grafs, Teoria de
Classificació AMS::05 Combinatorics::05C Graph theory
Àrees temàtiques de la UPC::Matemàtiques i estadística::Matemàtica discreta::Teoria de grafs
Descrição
Resumo:For a given random graph, a connected component that contains a finite fraction of the entire graph's vertices is called giant. The study of these components started with the \ER model, where it has been proven that removing the (unique) giant component from a random graph is essentially equivalent to another random graph in the same model with different known parameters. This is called the discrete duality principle. In this report we aim at presenting this principle in its historical \ER settings, and then to present more recent generalisations made on random graphs with given degree sequences.