Nonparametric factor analysis with beta process priors 论文

2009引用 271
Bayesian Methods and Mixture ModelsGene expression and cancer classificationStatistical Methods and Inference

详细信息

发表日期
2009-06-14
发表年份
2009

关键词

Bayesian Methods and Mixture ModelsGene expression and cancer classificationStatistical Methods and Inference

摘要

We propose a nonparametric extension to the factor analysis problem using a beta process prior. This beta process factor analysis (BP-FA) model allows for a dataset to be decomposed into a linear combination of a sparse set of factors, providing information on the underlying structure of the observations. As with the Dirichlet process, the beta process is a fully Bayesian conjugate prior, which allows for analytical posterior calculation and straightforward inference. We derive a variational Bayes inference algorithm and demonstrate the model on the MNIST digits and HGDP-CEPH cell line panel datasets. 1.