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.
Cite
Text
Han et al. "Variational Gaussian Copula Inference." International Conference on Artificial Intelligence and Statistics, 2016.Markdown
[Han et al. "Variational Gaussian Copula Inference." International Conference on Artificial Intelligence and Statistics, 2016.](https://mlanthology.org/aistats/2016/han2016aistats-variational/)BibTeX
@inproceedings{han2016aistats-variational,
title = {{Variational Gaussian Copula Inference}},
author = {Han, Shaobo and Liao, Xuejun and Dunson, David B. and Carin, Lawrence},
booktitle = {International Conference on Artificial Intelligence and Statistics},
year = {2016},
pages = {829-838},
url = {https://mlanthology.org/aistats/2016/han2016aistats-variational/}
}