I am a CNRS researcher at the Laboratoire de statistiques at CREST.

I hold a teaching position at ENSAE.


Bureau 3029
5, avenue Henry Le Chatelier
91120 Palaiseau

Tél.: +33(0)

PhD and Habilitation:

My PhD manuscript is available here and the slides of my PhD defense are available here.

My Habilitation manuscript is available here and the slides of my HDR defense are available here.

Some of my slides are available here:

  1. Optimal Aggregation via empirical risk minimization
  2. Aggregation methods in classification: optimality and adaptation
  3. Suboptimality of penalized empirical risk minimization in classification.
  4. Empirical risk minimization in linear regressiona dn phase recovery
  5. Learning sub-gaussian class. Minimaximality of ERM for high confidence.
  6. An aggregation procedure in classification. Applications to adaptivity.
  7. General oracle inequalities for ERM, regularized ERM and penalized ERM with applications to High-Dimensional data analysis


  1. Mark Fulk award” for the best student paper. 19th Annual Conference On Learning Theory, COLT06.
  2. Prix ROSEMONT/DEMASSIEUX” de la chancellerie des universités de Paris. Décembre 2008. For the best PhD thesis in Mathematics and its applications defended in (former) Paris University.


  1. My ha-index is 92; which is “too good” according to the official ha-index webpage. But it looks like there is no good explanation for the perfect interval [80, 85].
  2. About bibliometry. We were 19 from Université Paris-Est Marne-la-vallée to write the following two papers: Evgeny Abakumov, Anne Beaulieu, François Blanchard, Matthieu Fradelizi, Nathaël Gozlan, Bernard Host, Thiery Jeantheau, Magdalena Kobylanski, Guillaume Lecué, Miguel Martinez, Mathieu Meyer, Marie-Hélène Mourgues, Frédéric Portal, Francis Ribaud, Cyril Roberto, Pascal Romon, Julien Roth, Paul-Marie Samson, Pierre Vandekerkhove, Abdellah Youssfi.

*discussion *paper