Hierarchical Dirichlet processes and their applications: a survey

Jian Ying Zhou, Fei Yue Wang, Dajun Zeng

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Dirichlet processes are a type of stochastic processes widely used in nonparametric Bayesian models, especially in research that involves probabilistic graphical models. Over the past few years, significant effort has been made in the study of such processes, mainly due to their modeling flexibility and wide applicability. For instance, Dirichlet processes are capable of learning the number of clusters as well as the corresponding parameters of each cluster whereas other clustering or classification models usually are not able to. In this survey, we first introduce the definitions of Dirichlet processes. We then present Dirichlet process mixture models and their applications, and discuss in detail hierarchical Dirichlet processes (HDP), their roles in constructing other models, and examples of related applications in many important fields. Finally, we summarize recent developments in the study and applications of hierarchical Dirichlet processes and offer our remarks on future research.

Original languageEnglish (US)
Pages (from-to)389-407
Number of pages19
JournalZidonghua Xuebao/Acta Automatica Sinica
Volume37
Issue number4
DOIs
StatePublished - Apr 2011
Externally publishedYes

Fingerprint

Random processes

Keywords

  • Clustering
  • Dirichlet processes
  • Hierarchical Dirichlet processes (HDP)
  • Probabilistic graphical models

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Information Systems
  • Computer Graphics and Computer-Aided Design

Cite this

Hierarchical Dirichlet processes and their applications : a survey. / Zhou, Jian Ying; Wang, Fei Yue; Zeng, Dajun.

In: Zidonghua Xuebao/Acta Automatica Sinica, Vol. 37, No. 4, 04.2011, p. 389-407.

Research output: Contribution to journalArticle

@article{6452a8aca0f245a49b0ecc4a0feb2258,
title = "Hierarchical Dirichlet processes and their applications: a survey",
abstract = "Dirichlet processes are a type of stochastic processes widely used in nonparametric Bayesian models, especially in research that involves probabilistic graphical models. Over the past few years, significant effort has been made in the study of such processes, mainly due to their modeling flexibility and wide applicability. For instance, Dirichlet processes are capable of learning the number of clusters as well as the corresponding parameters of each cluster whereas other clustering or classification models usually are not able to. In this survey, we first introduce the definitions of Dirichlet processes. We then present Dirichlet process mixture models and their applications, and discuss in detail hierarchical Dirichlet processes (HDP), their roles in constructing other models, and examples of related applications in many important fields. Finally, we summarize recent developments in the study and applications of hierarchical Dirichlet processes and offer our remarks on future research.",
keywords = "Clustering, Dirichlet processes, Hierarchical Dirichlet processes (HDP), Probabilistic graphical models",
author = "Zhou, {Jian Ying} and Wang, {Fei Yue} and Dajun Zeng",
year = "2011",
month = "4",
doi = "10.3724/SP.J.1004.2011.00389",
language = "English (US)",
volume = "37",
pages = "389--407",
journal = "Zidonghua Xuebao/Acta Automatica Sinica",
issn = "0254-4156",
publisher = "Science Press",
number = "4",

}

TY - JOUR

T1 - Hierarchical Dirichlet processes and their applications

T2 - a survey

AU - Zhou, Jian Ying

AU - Wang, Fei Yue

AU - Zeng, Dajun

PY - 2011/4

Y1 - 2011/4

N2 - Dirichlet processes are a type of stochastic processes widely used in nonparametric Bayesian models, especially in research that involves probabilistic graphical models. Over the past few years, significant effort has been made in the study of such processes, mainly due to their modeling flexibility and wide applicability. For instance, Dirichlet processes are capable of learning the number of clusters as well as the corresponding parameters of each cluster whereas other clustering or classification models usually are not able to. In this survey, we first introduce the definitions of Dirichlet processes. We then present Dirichlet process mixture models and their applications, and discuss in detail hierarchical Dirichlet processes (HDP), their roles in constructing other models, and examples of related applications in many important fields. Finally, we summarize recent developments in the study and applications of hierarchical Dirichlet processes and offer our remarks on future research.

AB - Dirichlet processes are a type of stochastic processes widely used in nonparametric Bayesian models, especially in research that involves probabilistic graphical models. Over the past few years, significant effort has been made in the study of such processes, mainly due to their modeling flexibility and wide applicability. For instance, Dirichlet processes are capable of learning the number of clusters as well as the corresponding parameters of each cluster whereas other clustering or classification models usually are not able to. In this survey, we first introduce the definitions of Dirichlet processes. We then present Dirichlet process mixture models and their applications, and discuss in detail hierarchical Dirichlet processes (HDP), their roles in constructing other models, and examples of related applications in many important fields. Finally, we summarize recent developments in the study and applications of hierarchical Dirichlet processes and offer our remarks on future research.

KW - Clustering

KW - Dirichlet processes

KW - Hierarchical Dirichlet processes (HDP)

KW - Probabilistic graphical models

UR - http://www.scopus.com/inward/record.url?scp=79956327890&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79956327890&partnerID=8YFLogxK

U2 - 10.3724/SP.J.1004.2011.00389

DO - 10.3724/SP.J.1004.2011.00389

M3 - Article

AN - SCOPUS:79956327890

VL - 37

SP - 389

EP - 407

JO - Zidonghua Xuebao/Acta Automatica Sinica

JF - Zidonghua Xuebao/Acta Automatica Sinica

SN - 0254-4156

IS - 4

ER -