Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT
Zur Ablage hinzufügen
 Lokale TagsFreigabegeschichteDetailsÜbersicht
  Inference algorithms and learning theory for Bayesian sparse factor analysis

Rattray, M., Stegle, O., Sharp, K., & Winn, J. (2009). Inference algorithms and learning theory for Bayesian sparse factor analysis. Bristol, UK: Institute of Physics. doi:10.1088/1742-6596/197/1/012002.

Item is

 

einblenden: alle ausblenden: alle

Basisdaten

einblenden: ausblenden:
Datensatz-Permalink: https://hdl.handle.net/11858/00-001M-0000-0013-C2EA-9 Versions-Permalink: https://hdl.handle.net/21.11116/0000-0002-CB1E-9
Genre: Konferenzbeitrag

Externe Referenzen

einblenden:
ausblenden:
externe Referenz:
http://iopscience.iop.org/article/10.1088/1742-6596/197/1/012002/pdf (Verlagsversion)
Beschreibung:
-
OA-Status:

Urheber

einblenden:
ausblenden:
 Urheber:
Rattray, M, Autor
Stegle, O1, 2, Autor           
Sharp, K, Autor
Winn, J, Autor
Affiliations:
1Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497794              
2Former Research Group Machine Learning and Computational Biology, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_2528696              

Inhalt

einblenden:
ausblenden:
Schlagwörter: -
 Zusammenfassung: Bayesian sparse factor analysis has many applications; for example, it has been applied to the problem of inferring a sparse regulatory network from gene expression data. We describe a number of inference algorithms for Bayesian sparse factor analysis using a slab and spike mixture prior. These include well-established Markov chain Monte Carlo (MCMC) and variational Bayes (VB) algorithms as well as a novel hybrid of VB and Expectation Propagation (EP). For the case of a single latent factor we derive a theory for learning performance using the replica method. We compare the MCMC and VB/EP algorithm results with simulated data to the theoretical prediction. The results for MCMC agree closely with the theory as expected. Results for VB/EP are slightly sub-optimal but show that the new algorithm is effective for sparse inference. In large-scale problems MCMC is infeasible due to computational limitations and the VB/EP algorithm then provides a very useful computationally efficient alternative.

Details

einblenden:
ausblenden:
Sprache(n):
 Datum: 2009-12
 Publikationsstatus: Online veröffentlicht
 Seiten: -
 Ort, Verlag, Ausgabe: -
 Inhaltsverzeichnis: -
 Art der Begutachtung: -
 Identifikatoren: DOI: 10.1088/1742-6596/197/1/012002
BibTex Citekey: 6296
 Art des Abschluß: -

Veranstaltung

einblenden:
ausblenden:
Titel: International Workshop on Statistical-Mechanical Informatics 2009 (IW-SMI 2009)
Veranstaltungsort: Kyoto, Japan
Start-/Enddatum: 2009-09-13 - 2009-09-16

Entscheidung

einblenden:

Projektinformation

einblenden:

Quelle 1

einblenden:
ausblenden:
Titel: Journal of Physics: Conference Series
Genre der Quelle: Reihe
 Urheber:
Affiliations:
Ort, Verlag, Ausgabe: Bristol, UK : Institute of Physics
Seiten: - Band / Heft: 197 (1) Artikelnummer: 012002 Start- / Endseite: 1 - 10 Identifikator: -