logo Idiap Research Institute        
 [BibTeX] [Marc21]
Hierarchical Penalization
Type of publication: Idiap-RR
Citation: grandvalet:rr07-76
Number: Idiap-RR-76-2007
Year: 2007
Institution: IDIAP
Note: To appear in Advances in Neural Information Processing Systems 21 (NIPS 2007)
Abstract: Hierarchical penalization is a generic framework for incorporating prior information in the fitting of statistical models, when the explicative variables are organized in a hierarchical structure. The penalizer is a convex functional that performs soft selection at the group level, and shrinks variables within each group. This favors solutions with few leading terms in the final combination. The framework, originally derived for taking prior knowledge into account, is shown to be useful in linear regression, when several parameters are used to model the influence of one feature, or in kernel regression, for learning multiple kernels.
Userfields: ipdmembership={learning},
Keywords:
Projects Idiap
Authors Szafranski, Marie
Grandvalet, Yves
Morizet-Mahoudeaux, Pierre
Crossref by grandvalet:NIPS:2007
Added by: [UNK]
Total mark: 0
Attachments
  • grandvalet-idiap-rr-07-76.pdf
  • grandvalet-idiap-rr-07-76.ps.gz
Notes