Model selection by resampling penalization - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year : 2007

Model selection by resampling penalization

Sylvain Arlot


We present a new family of model selection algorithms based on the resampling heuristics. It can be used in several frameworks, do not require any knowledge about the unknown law of the data, and may be seen as a generalization of local Rademacher complexities and $V$-fold cross-validation. In the case example of least-square regression on histograms, we prove oracle inequalities, and that these algorithms are naturally adaptive to both the smoothness of the regression function and the variability of the noise level. Then, interpretating $V$-fold cross-validation in terms of penalization, we enlighten the question of choosing $V$. Finally, a simulation study illustrates the strength of resampling penalization algorithms against some classical ones, in particular with heteroscedastic data.
Fichier principal
Vignette du fichier
colt_court.pdf (268.38 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00125455 , version 1 (19-01-2007)
hal-00125455 , version 2 (22-01-2007)



Sylvain Arlot. Model selection by resampling penalization. 2007. ⟨hal-00125455v2⟩
107 View
90 Download



Gmail Facebook X LinkedIn More