jeudi 7 avril 2022
| Heures | événement | |
| 10:00 - 10:30 | Accueil (Amphi) | |
| 10:30 - 12:30 | Adversarial ML. (Amphi) | |
| 10:30 - 11:30 | › Generative Adversarial Networks: understanding optimality properties of Wasserstein GANs (WGANs) - Ugo Tanelian, CRITEO | |
| 11:30 - 12:30 | › Security threats in machine learning - Teddy Furon, Inria Rennes – Bretagne Atlantique | |
| 12:30 - 14:00 | Déjeuner | |
| 14:00 - 17:30 | Stochastic gradient descent (Amphi) | |
| 14:00 - 15:00 | › A review of nonconvex stochastic subgradient descent - Pascal Bianchi, Télécom Paris | |
| 15:00 - 16:00 | › Stochastic Gradient Descent with communication constraints and compression operators. - Aymeric Dieuleveut, Centre de Mathématiques Appliquées - Ecole Polytechnique | |
| 16:30 - 17:30 | › Stochastic gradient descents to online Newton algorithms - Antoine Godichon-Baggioni, UPMC |
vendredi 8 avril 2022
| Heures | événement | |
| 09:30 - 10:00 | Café | |
| 10:00 - 12:00 | Sampling (Amphi) | |
| 10:00 - 11:00 | › Adaptive Importance Sampling meets Mirror Descent: a Bias-variance tradeoff - Anna Korba, ENSAE | |
| 11:00 - 12:00 | › Non-Equilibrium Sampling - Alain Durmus - ENS (Paris-Saclay) | |
| 12:00 - 13:30 | Déjeuner | |
| 13:30 - 15:30 | EM algorithm | |
| 13:30 - 14:30 | › Properties of the stochastic approximation EM algorithm with mini-batch sampling - Estelle Kuhn - INRAE | |
| 14:30 - 15:30 | › Frugal Gaussian clustering of huge imbalanced datasets through a bin-marginal approach - Christine Keribin - Laboratoire de Mathématiques d'Orsay |