Variational Gaussian copula inference

Shaobo Han; Xuejun Liao; David B. Dunson; Lawrence Carin

Variational Gaussian copula inference

Shaobo Han, Xuejun Liao, David B. Dunson, Lawrence Carin

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models. For models with continuous and non-Gaussian hidden variables, we propose a semiparametric and automated variational Gaussian copula approach, in which the parametric Gaussian copula family is able to preserve multivariate posterior dependence, and the nonparametric transformations based on Bernstein polynomials provide ample flexibility in characterizing the univariate marginal posteriors.

Original language	English (US)
Title of host publication	Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, AISTATS 2016
Publisher	PMLR
Pages	829-838
Number of pages	10
State	Published - Jan 1 2016
Externally published	Yes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2021-02-09

Cite this

@inproceedings{38c8b17b5c3d4ebd90b331196215fbfc,

title = "Variational Gaussian copula inference",

abstract = "We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models. For models with continuous and non-Gaussian hidden variables, we propose a semiparametric and automated variational Gaussian copula approach, in which the parametric Gaussian copula family is able to preserve multivariate posterior dependence, and the nonparametric transformations based on Bernstein polynomials provide ample flexibility in characterizing the univariate marginal posteriors.",

author = "Shaobo Han and Xuejun Liao and Dunson, {David B.} and Lawrence Carin",

note = "Generated from Scopus record by KAUST IRTS on 2021-02-09",

year = "2016",

month = jan,

day = "1",

language = "English (US)",

pages = "829--838",

booktitle = "Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, AISTATS 2016",

publisher = "PMLR",

}

TY - GEN

T1 - Variational Gaussian copula inference

AU - Han, Shaobo

AU - Liao, Xuejun

AU - Dunson, David B.

AU - Carin, Lawrence

N1 - Generated from Scopus record by KAUST IRTS on 2021-02-09

PY - 2016/1/1

Y1 - 2016/1/1

N2 - We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models. For models with continuous and non-Gaussian hidden variables, we propose a semiparametric and automated variational Gaussian copula approach, in which the parametric Gaussian copula family is able to preserve multivariate posterior dependence, and the nonparametric transformations based on Bernstein polynomials provide ample flexibility in characterizing the univariate marginal posteriors.

AB - We utilize copulas to constitute a unified framework for constructing and optimizing variational proposals in hierarchical Bayesian models. For models with continuous and non-Gaussian hidden variables, we propose a semiparametric and automated variational Gaussian copula approach, in which the parametric Gaussian copula family is able to preserve multivariate posterior dependence, and the nonparametric transformations based on Bernstein polynomials provide ample flexibility in characterizing the univariate marginal posteriors.

UR - http://www.scopus.com/inward/record.url?scp=85040759228&partnerID=8YFLogxK

M3 - Conference contribution

SP - 829

EP - 838

BT - Proceedings of the 19th International Conference on Artificial Intelligence and Statistics, AISTATS 2016

PB - PMLR

ER -

Variational Gaussian copula inference

Abstract

Bibliographical note

Other files and links

Fingerprint

Cite this