Résumé : We consider the problem of estimating a function in
logistic regression model. We propose to estimate this function
by a sparse approximation build as a linear combination of
elements of a given dictionary of functions. This sparse
approximation is selected by the Lasso or Group Lasso procedure. In
this context, we state non asymptotic oracle inequalities for Lasso
and Group Lasso under restricted eigenvalues assumption as introduced
in Bickel et al. (2009). Those theoretical results are illustrated
through a simulation study.

Cet exposé se tiendra en salle C20-13, 20ème étage, Université
Paris 1, Centre Pierre Mendès-France, 90 rue de Tolbiac, 75013 Paris
(métro : Olympiades).