Document Detail


l(p)-l(q) Penalty for Sparse Linear and Sparse Multiple Kernel Multitask Learning.
MedLine Citation:
PMID:  21813358     Owner:  NLM     Status:  In-Data-Review    
Abstract/OtherAbstract:
Recently, there has been much interest around multitask learning (MTL) problem with the constraints that tasks should share a common sparsity profile. Such a problem can be addressed through a regularization framework where the regularizer induces a joint-sparsity pattern between task decision functions. We follow this principled framework and focus on l(p)-l(q) (with 0 ≤ p ≤ 1 and 1 ≤ q ≤ 2) mixed norms as sparsity-inducing penalties. Our motivation for addressing such a larger class of penalty is to adapt the penalty to a problem at hand leading thus to better performances and better sparsity pattern. For solving the problem in the general multiple kernel case, we first derive a variational formulation of the l(1)-l(q) penalty which helps us in proposing an alternate optimization algorithm. Although very simple, the latter algorithm provably converges to the global minimum of the l(1)-l(q) penalized problem. For the linear case, we extend existing works considering accelerated proximal gradient to this penalty. Our contribution in this context is to provide an efficient scheme for computing the l(1)-l(q) proximal operator. Then, for the more general case, when , we solve the resulting nonconvex problem through a majorization-minimization approach. The resulting algorithm is an iterative scheme which, at each iteration, solves a weighted l(1)-l(q) sparse MTL problem. Empirical evidences from toy dataset and real-word datasets dealing with brain-computer interface single-trial electroencephalogram classification and protein subcellular localization show the benefit of the proposed approaches and algorithms.
Authors:
Alain Rakotomamonjy; Rémi Flamary; Gilles Gasso; Stéphane Canu
Related Documents :
12182798 - Design guidelines for interactive multimedia learning environments to promote social in...
21706758 - Introductory course based on a single problem: learning nucleic acid biochemistry from ...
3626898 - Content coverage in problem-based learning.
15369908 - Group problems in problem-based learning.
2317308 - Reducing the spread of hiv infection in sub-saharan africa: some demographic and econom...
16897378 - Physical illness, pain, and problem behavior in minimally verbal people with developmen...
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  IEEE transactions on neural networks / a publication of the IEEE Neural Networks Council     Volume:  22     ISSN:  1941-0093     ISO Abbreviation:  IEEE Trans Neural Netw     Publication Date:  2011 Aug 
Date Detail:
Created Date:  2011-08-04     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101211035     Medline TA:  IEEE Trans Neural Netw     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1307-20     Citation Subset:  IM    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Convergence of cyclic and almost-cyclic learning with momentum for feedforward neural networks.
Next Document:  Adaptive Neural Output Feedback Controller Design With Reduced-Order Observer for a Class of Uncerta...