Show simple item record

dc.contributor.author Yoshida, R
dc.contributor.author West, M
dc.coverage.spatial United States
dc.date.accessioned 2011-06-21T17:32:27Z
dc.date.issued 2010-05-01
dc.identifier http://www.ncbi.nlm.nih.gov/pubmed/20890391
dc.identifier.citation J Mach Learn Res, 2010, 11 pp. 1771 - 1798
dc.identifier.issn 1532-4435
dc.identifier.uri http://hdl.handle.net/10161/4635
dc.description.abstract We describe a class of sparse latent factor models, called graphical factor models (GFMs), and relevant sparse learning algorithms for posterior mode estimation. Linear, Gaussian GFMs have sparse, orthogonal factor loadings matrices, that, in addition to sparsity of the implied covariance matrices, also induce conditional independence structures via zeros in the implied precision matrices. We describe the models and their use for robust estimation of sparse latent factor structure and data/signal reconstruction. We develop computational algorithms for model exploration and posterior mode search, addressing the hard combinatorial optimization involved in the search over a huge space of potential sparse configurations. A mean-field variational technique coupled with annealing is developed to successively generate "artificial" posterior distributions that, at the limiting temperature in the annealing schedule, define required posterior modes in the GFM parameter space. Several detailed empirical studies and comparisons to related approaches are discussed, including analyses of handwritten digit image and cancer gene expression data.
dc.format.extent 1771 - 1798
dc.language ENG
dc.language.iso en_US en_US
dc.relation.ispartof J Mach Learn Res
dc.title Bayesian Learning in Sparse Graphical Factor Models via Variational Mean-Field Annealing.
dc.title.alternative en_US
dc.type Journal Article
dc.description.version Version of Record en_US
duke.date.pubdate 2010-5-0 en_US
duke.description.endpage 1798 en_US
duke.description.issue en_US
duke.description.startpage 1771 en_US
duke.description.volume 11 en_US
dc.relation.journal Journal of Machine Learning Research en_US
pubs.author-url http://www.ncbi.nlm.nih.gov/pubmed/20890391
pubs.organisational-group /Duke
pubs.organisational-group /Duke/School of Medicine
pubs.organisational-group /Duke/School of Medicine/Institutes and Centers
pubs.organisational-group /Duke/School of Medicine/Institutes and Centers/Duke Cancer Institute
pubs.organisational-group /Duke/Trinity College of Arts & Sciences
pubs.organisational-group /Duke/Trinity College of Arts & Sciences/Statistical Science
pubs.publication-status Published
pubs.volume 11

Files in this item

This item appears in the following Collection(s)

Show simple item record