Bayesian Group Factor Analysis

File Description SizeFormat 
virtanen12.pdfPublished version444.79 kBAdobe PDFView/Open
Title: Bayesian Group Factor Analysis
Authors: Virtanen, S
Klami, A
Khan, SA
Kaski, S
Item Type: Conference Paper
Abstract: We introduce a factor analysis model that summarizes the dependencies between observed variable groups, instead of dependencies between individual variables as standard factor analysis does. A group may correspond to one view of the same set of objects, one of many data sets tied by co-occurrence, or a set of alternative variables collected from statistics tables to measure one property of interest. We show that by assuming group-wise sparse factors, active in a subset of the sets, the variation can be decomposed into factors explaining relationships between the sets and factors explaining away set-specific variation. We formulate the assumptions in a Bayesian model which provides the factors, and apply the model to two data analysis tasks, in neuroimaging and chemical systems biology.
Issue Date: 31-Jan-2012
Date of Acceptance: 1-Jan-2012
URI: http://hdl.handle.net/10044/1/52920
Start Page: 1269
End Page: 1277
Journal / Book Title: Proceedings of the 15th AISTATS
Copyright Statement: Copyright 2012 by the authors.
Conference Name: 15th International Conference on Artificial Intelligence and Statistics (AISTATS)
Keywords: stat.ML
Notes: 9 pages, 5 figures
Publication Status: Published
Start Date: 2012-04-21
Finish Date: 2012-04-23
Conference Place: La Palma, Canary Islands
Appears in Collections:Mathematics
Statistics
Faculty of Natural Sciences



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Creative Commonsx