文章基本信息

标题：A General Method for Robust Bayesian Modeling
本地全文：下载
作者：Chong Wang ; David M. Blei
期刊名称：Bayesian Analysis
印刷版ISSN：1931-6690
电子版ISSN：1936-0975
出版年度：2018
卷号：13
期号：4
页码：1163-1191
DOI：10.1214/17-BA1090
语种：English
出版社：International Society for Bayesian Analysis
摘要：Robust Bayesian models are appealing alternatives to standard models, providing protection from data that contains outliers or other departures from the model assumptions. Historically, robust models were mostly developed on a case-by-case basis; examples include robust linear regression, robust mixture models, and bursty topic models. In this paper we develop a general approach to robust Bayesian modeling. We show how to turn an existing Bayesian model into a robust model, and then develop a generic computational strategy for it. We use our method to study robust variants of several models, including linear regression, Poisson regression, logistic regression, and probabilistic topic models. We discuss the connections between our methods and existing approaches, especially empirical Bayes and James–Stein estimation.
关键词：robust statistics; empirical Bayes; probabilistic models; variational inference; expectation-maximization; generalized linear models; topic models.