Covariate-dependent dictionary learning and sparse coding

A dependent hierarchical beta process (dHBP) is developed as a prior for data that may be represented in terms of a sparse set of latent features (dictionary elements), with covariate dependent feature usage. The dHBP is applicable to general covariates and data models, imposing that signals with si...

Full description

Saved in:
Bibliographic Details
Published in2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 5824 - 5827
Main Authors Mingyuan Zhou, Hongxia Yang, Sapiro, Guillermo, Dunson, David, Carin, Lawrence
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2011
Subjects
Atomic measurements
Bayesian
Bismuth
covariates
Dictionaries
dictionary learning
Face
Information processing
Kernel
Manifolds
sparse coding
Online AccessGet full text

Cover

More Information
Summary:A dependent hierarchical beta process (dHBP) is developed as a prior for data that may be represented in terms of a sparse set of latent features (dictionary elements), with covariate dependent feature usage. The dHBP is applicable to general covariates and data models, imposing that signals with similar covariates are likely to be manifested in terms of similar features. As an application, we consider the simultaneous sparse modeling of multiple images, with the covariate of a given image linked to its similarity to all other images (as applied in manifold learning). Efficient inference is performed using hybrid Gibbs, Metropolis-Hastings and slice sampling.
ISBN:9781457705380
1457705389
ISSN:1520-6149
DOI:10.1109/ICASSP.2011.5947685