Sciweavers

Share
IDA
2003
Springer

Regularization Methods for Additive Models

10 years 5 months ago
Regularization Methods for Additive Models
This paper tackles the problem of model complexity in the context of additive models. Several methods have been proposed to estimate smoothing parameters, as well as to perform variable selection. Nevertheless, these procedures are inefficient or computationally expensive in high dimension. Also, the lasso technique has been adapted to additive models, however its experimental performance has not been analyzed. We propose a modified lasso for additive models, improving variable selection. A benchmark is also developed, to examine its practical behavior, comparing it with forward selection. Our simulation studies suggest ability to carry out model selection of the proposed method. The lasso technique shows up better than forward in the most complex situations. The computing time of modified lasso is considerably smaller since it does not depend on the number of relevant variables.
Marta Avalos, Yves Grandvalet, Christophe Ambroise
Added 07 Jul 2010
Updated 07 Jul 2010
Type Conference
Year 2003
Where IDA
Authors Marta Avalos, Yves Grandvalet, Christophe Ambroise
Comments (0)
books