Skip to main content

Nonparametric factor analysis with beta process priors

Publication ,  Journal Article
Paisley, J; Carin, L
Published in: ACM International Conference Proceeding Series
September 15, 2009

We propose a nonparametric extension to the factor analysis problem using a beta process prior. This beta process factor analysis (BPFA) 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. Copyright 2009.

Duke Scholars

Altmetric Attention Stats
Dimensions Citation Stats

Published In

ACM International Conference Proceeding Series

DOI

Publication Date

September 15, 2009

Volume

382
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Paisley, J., & Carin, L. (2009). Nonparametric factor analysis with beta process priors. ACM International Conference Proceeding Series, 382. https://doi.org/10.1145/1553374.1553474
Paisley, J., and L. Carin. “Nonparametric factor analysis with beta process priors.” ACM International Conference Proceeding Series 382 (September 15, 2009). https://doi.org/10.1145/1553374.1553474.
Paisley J, Carin L. Nonparametric factor analysis with beta process priors. ACM International Conference Proceeding Series. 2009 Sep 15;382.
Paisley, J., and L. Carin. “Nonparametric factor analysis with beta process priors.” ACM International Conference Proceeding Series, vol. 382, Sept. 2009. Scopus, doi:10.1145/1553374.1553474.
Paisley J, Carin L. Nonparametric factor analysis with beta process priors. ACM International Conference Proceeding Series. 2009 Sep 15;382.

Published In

ACM International Conference Proceeding Series

DOI

Publication Date

September 15, 2009

Volume

382