Group Lasso with Overlaps: the Latent Group Lasso approach

Abstract : We study a norm for structured sparsity which leads to sparse linear predictors whose supports are unions of prede ned overlapping groups of variables. We call the obtained formulation latent group Lasso, since it is based on applying the usual group Lasso penalty on a set of latent variables. A detailed analysis of the norm and its properties is presented and we characterize conditions under which the set of groups associated with latent variables are correctly identi ed. We motivate and discuss the delicate choice of weights associated to each group, and illustrate this approach on simulated data and on the problem of breast cancer prognosis from gene expression data.
Document type :
Reports
[Research Report] 2011, pp.60


https://hal.inria.fr/inria-00628498
Contributor : Guillaume Obozinski <>
Submitted on : Monday, October 3, 2011 - 2:49:33 PM
Last modification on : Monday, October 3, 2011 - 6:49:47 PM

Files

lglasso-hal.pdf
fileSource_public_author

Identifiers

  • HAL Id : inria-00628498, version 1
  • ARXIV : 1110.0413

Collections

Citation

Guillaume Obozinski, Laurent Jacob, Jean-Philippe Vert. Group Lasso with Overlaps: the Latent Group Lasso approach. [Research Report] 2011, pp.60. <inria-00628498>

Export

Share

Metrics

Consultation de
la notice

497

Téléchargement du document

550