Continuum limit of critical inhomogeneous random graphs

Shankar Bhamidi, Sanchayan Sen, Xuan Wang

Research output: Contribution to journalArticle

  • 2 Citations

Abstract

The last few years have witnessed tremendous interest in understanding the structure as well as the behavior of dynamics for inhomogeneous random graph models to gain insight into real-world systems. In this study we analyze the maximal components at criticality of one famous class of such models, the rank-one inhomogeneous random graph model (Norros and Reittu, Adv Appl Probab 38(1):59–75, 2006; Bollobás et al., Random Struct Algorithms 31(1):3–122, 2007, Section 16.4). Viewing these components as measured random metric spaces, under finite moment assumptions for the weight distribution, we show that the components in the critical scaling window with distances scaled by n- 1 / 3 converge in the Gromov–Haussdorf–Prokhorov metric to rescaled versions of the limit objects identified for the Erdős–Rényi random graph components at criticality in Addario-Berry et al. (Probab. Theory Related Fields, 152(3–4):367–406, 2012). A key step is the construction of connected components of the random graph through an appropriate tilt of a fundamental class of random trees called p-trees (Camarri and Pitman, Electron. J. Probab 5(2):1–18, 2000; Aldous et al., Probab Theory Related Fields 129(2):182–218, 2004). This is the first step in rigorously understanding the scaling limits of objects such as the minimal spanning tree and other strong disorder models from statistical physics (Braunstein et al., Phys Rev Lett 91(16):168701, 2003) for such graph models. By asymptotic equivalence (Janson, Random Struct Algorithms 36(1):26–45, 2010), the same results are true for the Chung–Lu model (Chung and Lu, Proc Natl Acad Sci 99(25):15879–15882, 2002; Chung and Lu, Ann Combin 6(2):125–145, 2002; Chung and Lu, Complex graphs and networks, 2006) and the Britton–Deijfen–Martin–Löf model (Britton et al., J Stat Phys 124(6):1377–1397, 2006). A crucial ingredient of the proof of independent interest are tail bounds for the height of p-trees. The techniques developed in this paper form the main technical bedrock for the general program developed in Bhamidi et al. (Scaling limits of random graph models at criticality: Universality and the basin of attraction of the Erdős–Rényi random graph. arXiv preprint, 2014) for proving universality of the continuum scaling limits in the critical regime for a wide array of other random graph models including the configuration model and inhomogeneous random graphs with general kernels (Bollobás et al., Random Struct Algorithms 31(1):3–122, 2007).

LanguageEnglish (US)
Pages565-641
Number of pages77
JournalProbability Theory and Related Fields
Volume169
Issue number1-2
DOIs
StatePublished - Oct 1 2017

Fingerprint

Continuum Limit
Random Graphs
Graph Model
Scaling Limit
Criticality
Universality
Prokhorov Metric
Minimal Spanning Tree
Asymptotic Equivalence
Random Trees
Weight Distribution
Model
Random graphs
Basin of Attraction
Statistical Physics
Tilt
Connected Components
Metric space
Disorder
Tail

Keywords

  • Branching processes
  • Continuum random tree
  • Critical random graphs
  • Multiplicative coalescent
  • p-trees
  • Scaling limits

ASJC Scopus subject areas

  • Analysis
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Continuum limit of critical inhomogeneous random graphs. / Bhamidi, Shankar; Sen, Sanchayan; Wang, Xuan.

In: Probability Theory and Related Fields, Vol. 169, No. 1-2, 01.10.2017, p. 565-641.

Research output: Contribution to journalArticle

Bhamidi, Shankar ; Sen, Sanchayan ; Wang, Xuan. / Continuum limit of critical inhomogeneous random graphs. In: Probability Theory and Related Fields. 2017 ; Vol. 169, No. 1-2. pp. 565-641
@article{16395c90c6204d2e8cd6de4ceced8381,
title = "Continuum limit of critical inhomogeneous random graphs",
abstract = "The last few years have witnessed tremendous interest in understanding the structure as well as the behavior of dynamics for inhomogeneous random graph models to gain insight into real-world systems. In this study we analyze the maximal components at criticality of one famous class of such models, the rank-one inhomogeneous random graph model (Norros and Reittu, Adv Appl Probab 38(1):59–75, 2006; Bollob{\'a}s et al., Random Struct Algorithms 31(1):3–122, 2007, Section 16.4). Viewing these components as measured random metric spaces, under finite moment assumptions for the weight distribution, we show that the components in the critical scaling window with distances scaled by n- 1 / 3 converge in the Gromov–Haussdorf–Prokhorov metric to rescaled versions of the limit objects identified for the Erdős–R{\'e}nyi random graph components at criticality in Addario-Berry et al. (Probab. Theory Related Fields, 152(3–4):367–406, 2012). A key step is the construction of connected components of the random graph through an appropriate tilt of a fundamental class of random trees called p-trees (Camarri and Pitman, Electron. J. Probab 5(2):1–18, 2000; Aldous et al., Probab Theory Related Fields 129(2):182–218, 2004). This is the first step in rigorously understanding the scaling limits of objects such as the minimal spanning tree and other strong disorder models from statistical physics (Braunstein et al., Phys Rev Lett 91(16):168701, 2003) for such graph models. By asymptotic equivalence (Janson, Random Struct Algorithms 36(1):26–45, 2010), the same results are true for the Chung–Lu model (Chung and Lu, Proc Natl Acad Sci 99(25):15879–15882, 2002; Chung and Lu, Ann Combin 6(2):125–145, 2002; Chung and Lu, Complex graphs and networks, 2006) and the Britton–Deijfen–Martin–L{\"o}f model (Britton et al., J Stat Phys 124(6):1377–1397, 2006). A crucial ingredient of the proof of independent interest are tail bounds for the height of p-trees. The techniques developed in this paper form the main technical bedrock for the general program developed in Bhamidi et al. (Scaling limits of random graph models at criticality: Universality and the basin of attraction of the Erdős–R{\'e}nyi random graph. arXiv preprint, 2014) for proving universality of the continuum scaling limits in the critical regime for a wide array of other random graph models including the configuration model and inhomogeneous random graphs with general kernels (Bollob{\'a}s et al., Random Struct Algorithms 31(1):3–122, 2007).",
keywords = "Branching processes, Continuum random tree, Critical random graphs, Multiplicative coalescent, p-trees, Scaling limits",
author = "Shankar Bhamidi and Sanchayan Sen and Xuan Wang",
year = "2017",
month = "10",
day = "1",
doi = "10.1007/s00440-016-0737-x",
language = "English (US)",
volume = "169",
pages = "565--641",
journal = "Probability Theory and Related Fields",
issn = "0178-8051",
publisher = "Springer New York",
number = "1-2",

}

TY - JOUR

T1 - Continuum limit of critical inhomogeneous random graphs

AU - Bhamidi,Shankar

AU - Sen,Sanchayan

AU - Wang,Xuan

PY - 2017/10/1

Y1 - 2017/10/1

N2 - The last few years have witnessed tremendous interest in understanding the structure as well as the behavior of dynamics for inhomogeneous random graph models to gain insight into real-world systems. In this study we analyze the maximal components at criticality of one famous class of such models, the rank-one inhomogeneous random graph model (Norros and Reittu, Adv Appl Probab 38(1):59–75, 2006; Bollobás et al., Random Struct Algorithms 31(1):3–122, 2007, Section 16.4). Viewing these components as measured random metric spaces, under finite moment assumptions for the weight distribution, we show that the components in the critical scaling window with distances scaled by n- 1 / 3 converge in the Gromov–Haussdorf–Prokhorov metric to rescaled versions of the limit objects identified for the Erdős–Rényi random graph components at criticality in Addario-Berry et al. (Probab. Theory Related Fields, 152(3–4):367–406, 2012). A key step is the construction of connected components of the random graph through an appropriate tilt of a fundamental class of random trees called p-trees (Camarri and Pitman, Electron. J. Probab 5(2):1–18, 2000; Aldous et al., Probab Theory Related Fields 129(2):182–218, 2004). This is the first step in rigorously understanding the scaling limits of objects such as the minimal spanning tree and other strong disorder models from statistical physics (Braunstein et al., Phys Rev Lett 91(16):168701, 2003) for such graph models. By asymptotic equivalence (Janson, Random Struct Algorithms 36(1):26–45, 2010), the same results are true for the Chung–Lu model (Chung and Lu, Proc Natl Acad Sci 99(25):15879–15882, 2002; Chung and Lu, Ann Combin 6(2):125–145, 2002; Chung and Lu, Complex graphs and networks, 2006) and the Britton–Deijfen–Martin–Löf model (Britton et al., J Stat Phys 124(6):1377–1397, 2006). A crucial ingredient of the proof of independent interest are tail bounds for the height of p-trees. The techniques developed in this paper form the main technical bedrock for the general program developed in Bhamidi et al. (Scaling limits of random graph models at criticality: Universality and the basin of attraction of the Erdős–Rényi random graph. arXiv preprint, 2014) for proving universality of the continuum scaling limits in the critical regime for a wide array of other random graph models including the configuration model and inhomogeneous random graphs with general kernels (Bollobás et al., Random Struct Algorithms 31(1):3–122, 2007).

AB - The last few years have witnessed tremendous interest in understanding the structure as well as the behavior of dynamics for inhomogeneous random graph models to gain insight into real-world systems. In this study we analyze the maximal components at criticality of one famous class of such models, the rank-one inhomogeneous random graph model (Norros and Reittu, Adv Appl Probab 38(1):59–75, 2006; Bollobás et al., Random Struct Algorithms 31(1):3–122, 2007, Section 16.4). Viewing these components as measured random metric spaces, under finite moment assumptions for the weight distribution, we show that the components in the critical scaling window with distances scaled by n- 1 / 3 converge in the Gromov–Haussdorf–Prokhorov metric to rescaled versions of the limit objects identified for the Erdős–Rényi random graph components at criticality in Addario-Berry et al. (Probab. Theory Related Fields, 152(3–4):367–406, 2012). A key step is the construction of connected components of the random graph through an appropriate tilt of a fundamental class of random trees called p-trees (Camarri and Pitman, Electron. J. Probab 5(2):1–18, 2000; Aldous et al., Probab Theory Related Fields 129(2):182–218, 2004). This is the first step in rigorously understanding the scaling limits of objects such as the minimal spanning tree and other strong disorder models from statistical physics (Braunstein et al., Phys Rev Lett 91(16):168701, 2003) for such graph models. By asymptotic equivalence (Janson, Random Struct Algorithms 36(1):26–45, 2010), the same results are true for the Chung–Lu model (Chung and Lu, Proc Natl Acad Sci 99(25):15879–15882, 2002; Chung and Lu, Ann Combin 6(2):125–145, 2002; Chung and Lu, Complex graphs and networks, 2006) and the Britton–Deijfen–Martin–Löf model (Britton et al., J Stat Phys 124(6):1377–1397, 2006). A crucial ingredient of the proof of independent interest are tail bounds for the height of p-trees. The techniques developed in this paper form the main technical bedrock for the general program developed in Bhamidi et al. (Scaling limits of random graph models at criticality: Universality and the basin of attraction of the Erdős–Rényi random graph. arXiv preprint, 2014) for proving universality of the continuum scaling limits in the critical regime for a wide array of other random graph models including the configuration model and inhomogeneous random graphs with general kernels (Bollobás et al., Random Struct Algorithms 31(1):3–122, 2007).

KW - Branching processes

KW - Continuum random tree

KW - Critical random graphs

KW - Multiplicative coalescent

KW - p-trees

KW - Scaling limits

UR - http://www.scopus.com/inward/record.url?scp=84984813430&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84984813430&partnerID=8YFLogxK

U2 - 10.1007/s00440-016-0737-x

DO - 10.1007/s00440-016-0737-x

M3 - Article

VL - 169

SP - 565

EP - 641

JO - Probability Theory and Related Fields

T2 - Probability Theory and Related Fields

JF - Probability Theory and Related Fields

SN - 0178-8051

IS - 1-2

ER -