This week will be devoted to statistical learning from both the theoretical and applied perspectives. Statistical learning theory has been developed in the 70’s and brought a great revival in statistics. On the one hand, the development of computer science and tools allowed massive data collection, and implementation of powerful algorithms which are often memory and computation time consuming. On the other hand, classical asymptotic theory used to prove the efficiency of estimation methods in modeling and prediction was limited by dimensionality problems.
         The approaches developed in statistical learning helped to face some new challenges such as the curse of dimensionality, small sample size, and now massive  datasets. Data mining, feature selection, and more recently Big data appeared as specialized approaches for massive datasets modeling and analysis.
         In addition to the theoretical developments (non asymptotic theory), many algorithms emerged in statistics and computer science to meet these new needs which arised in many areas such as bioinformatics, social sciences, medicine or telecommunications.
         This week aims to bring together specialists from statistical learning working on advanced techniques and coming from different fields, mainly statistics, but also computer science, social science and bioinformatics.
The main topics of interest include:

• Supervised learning
• Unsupervised learning: clustering and density estimation
• Model selection
• Big data

Scientific Committee

Gérard Biau (Université Pierre-et-Marie-Curie)
Pascal Massart (Université Paris-Sud)
Vincent Rivoirard (Université Paris-Dauphine)

Organizing Committee

Badih Ghattas (Aix-Marseille Université)
Liva Ralaivola (Aix-Marseille Université)

Speakers

• Francis Bach (ENS Paris, France)

Large-scale machine learning and convex optimization

• Philippe Besse (INSA Toulouse)

Apprentissage et données massives

• Gilles Blanchard (Potsdam University)

Is adaptive early stopping possible in statistical inverse problems?

• Sébastien Bubeck (Princeton University & Microsoft Seattle)

Entropy, geometry, and a CLT for Wishart matrices

• Peter Buhlmann (ETH Zurich)

The power of heterogeneous large-scale data for high-dimensional causal inference

• Stéphane Canu (INSA Rouen)

Mixed integer programming for sparse and non convex machine learning

• Stéphane Chrétien (National Physical Laboratory, Teddington, UK)

A Lagrangian viewpoint on Robust PCA

• Emilie Devijver (Orsay-Belgique)

Block-diagonal covariance selection for high-dimensional Gaussian
graphical models

• Stéphane Gaïffas (Ecole polytechnique)

Statistical learning with Hawkes processes and new matrix concentration inequalities

• Pierre Geurts (Université de Liège)

Random forests variable importances: towards a better understanding and large-scale feature selection

• Claire Lacour (Université Paris-Sud)

About the Goldenshluger-Lepski methodology for bandwidth selection

• Matthieu Lerasle (Université Nice-Sophia-Antipolis)

Sub-Gaussian mean estimators

• Clément Levrard (Université Paris Diderot)

Reconstruction simpliciale de variétés via l’estimation des plans tangents

•  Sébastien Loustau (Université d’Angers)

Quantization, Learning and Games with OPAC

• Stéphane Mallat (Ecole polytechnique)

Understanding (or not) Deep Neural Networks

• André Mas (Université de Montpellier)

Eigenvalue-free risk bounds for PCA projectors

• Patricia Reynaud-Bouret (Université Nice-Sophia-Antipolis)

Estimation  of  local  independence  graphs  via  Hawkes  processes  and link with the  functional neuronal connectivity

• Eric Sibony (Telecom Paris Tech)

A novel multi resolution framework for the statistical analysis of ranking data

• Gilles Stoltz (CNRS, HEC Paris, France)

Robust sequential learning with applications to the forecasting of electricity consumption and of exchange rates

• Alexandre Tsybakov (CREST-ENSAE)

Oracle inequalities for network models and sparse graphon estimation