Robustness and computational efficiency of algorithms in statistical learning
RENCONTRE DE STATISTIQUES MATHEMATIQUES
Noisy and corrupted measurements, commonly encountered as a result of unsupervised, automated data collection process, require new algorithms that produce reliable outcomes in this challenging framework. Discussing the influence of outliers on statistical procedures, P. Huber observed that « …the naturally occurring deviations from the idealized model are large enough to render meaningless the traditional asymptotic optimality theory. » It is known that heuristic outlier removal procedures are bias-producing and lack rigorous justification (or require strong assumptions), as it is sometimes impossible to determine if an extreme observation appears due to an error, or is a feature of the data-generating mechanism. This motivates the study of robust estimators in the context of statistical learning theory.
Our goal is to encourage collaboration and knowledge sharing between theoretical computer scientists and mathematical statisticians by bringing them together at Luminy. Both communities possess unique visions, skills and expertise, we hope that merging these strength will advance the field and bring us closer to solving some of the key challenges. Organizing Committee & Scientific Committee
Cristina Butucea (CREST, ENSAE, Université Paris-Est Marne-la-Vallée) |
Discussion rooms will be available to registered conference participants via the link below. Your access code to the rooms is provided by the organizers.
This page is accessible with a password issued by CIRM.
Use of Big Blue Button
|
Chao Gao (University of Chicago) Statistical Optimality and Algorithms for Top-K and Total Ranking
TALKS
Felix Abramovich (Tel Aviv University)
High-dimensional classification by sparse logistic regression Pierre Bellec (Rutgers University) Clément Berenfeld (Université Paris-Dauphine) Thomas Berrett (University of Warwick) Natalia Bochkina (University of Edinburgh) Alexandra Carpentier (OvGU Magdeburg) Julien Chhor (CREST-ENSAE) Fabienne Comte (Université Paris Descartes) Alexis Derumigny (University of Twente) Motonobu Kanagawa (EURECOM) Avetik Karagulyan (ENSAE/CREST) Clément Marteau (Université Lyon 1) |
Mohamed Simo Ndaoud (University of Southern California)
Robust and efficient mean estimation: approach based on the properties of self-normalized sums Tuan-Binh Nguyen (Universite Paris-Sud (Paris-Saclay) Marianna Pensky (University of Central Florida) Vianney Perchet (ENSAE & Criteo AI Lab) Philippe Rigollet (MIT) Vincent Rivoirard (Université Paris-Dauphine) Angelika Rohde (Albert-Ludwigs-University Freiburg) Richard Samworth (University of Cambridge) Johannes Schmidt-Hieber (University of Twente) Suzanne Sigalla (CREST-ENSAE) Zoltan Szabo (Ecole Polytechnique) Nicolas Verzelen (INRAE Montpellier) Nikita Zhivotovsky (Google Research) |