4.7 Article

Posterior Asymptotics for Boosted Hierarchical Dirichlet Process Mixtures

期刊

出版社

MICROTOME PUBL

关键词

Bayesian asymptotics; Dirichlet process; hierarchical process; nonparametric density estimation; partial exchangeability; Pitman-Yor process; posterior contraction rates

向作者/读者索取更多资源

Bayesian hierarchical models are powerful tools for learning common latent features across multiple data sources. This study establishes theoretical guarantees for recovering the true data generating process in the Hierarchical Dirichlet Process (HDP) or a generalization of the HDP. The posterior contraction rates are affected by the relationship between sample sizes.
Bayesian hierarchical models are powerful tools for learning common latent features across multiple data sources. The Hierarchical Dirichlet Process (HDP) is invoked when the number of latent components is a priori unknown. While there is a rich literature on finite sample properties and performance of hierarchical processes, the analysis of their frequentist posterior asymptotic properties is still at an early stage. Here we establish theoretical guarantees for recovering the true data generating process when the data are modeled as mixtures over the HDP or a generalization of the HDP, which we term boosted because of the faster growth in the number of discovered latent features. By extending Schwartz's theory to partially exchangeable sequences we show that posterior contraction rates are crucially affected by the relationship between the sample sizes corresponding to the different groups. The effect varies according to the smoothness level of the true data distributions. In the supersmooth case, when the generating densities are Gaussian mixtures, we recover the parametric rate up to a logarithmic factor, provided that the sample sizes are related in a polynomial fashion. Under ordinary smoothness assumptions more caution is needed as a polynomial deviation in the sample sizes could drastically deteriorate the convergence to the truth.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据