Skip to main navigation Skip to search Skip to main content

Bayesian supervised learning with non-Gaussian latent variables

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

We describe a Bayesian learning scheme for the hierarchal Bayesian linear model, which is based on the Gaussian scale mixture (GSM) modeling of the distribution of the latent variable. The proposed method takes advantage of the hierarchal Gaussian structure for a simple Monte-Carlo sampling algorithm. Particularly, with a single hidden scale parameter controlling the distribution of the latent variables, it leads to an efficient algorithm without explicit matrix inversion.

Original languageEnglish
Title of host publication2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Proceedings
Pages659-663
Number of pages5
DOIs
StatePublished - 2013
Event2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Beijing, China
Duration: Jul 6 2013Jul 10 2013

Publication series

Name2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013 - Proceedings

Conference

Conference2013 IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2013
Country/TerritoryChina
CityBeijing
Period07/6/1307/10/13

Keywords

  • Bayesian learning
  • Gaussian scale mixtures
  • latent variable models

Fingerprint

Dive into the research topics of 'Bayesian supervised learning with non-Gaussian latent variables'. Together they form a unique fingerprint.

Cite this