Collection Statistics & Computer Science Day for Data Science at Paris-Saclay 2021

The aim of this statistical and computer science day is to bring together mathematicians and computer scientists around some talks on recent results from statistics and machine learning. Various topics in machine learning, optimization, deep learning, or optimal transport will be presented.

Organizer(s) Thanh Mai Pham Ngoc (LMO) and Charles Soussen (L2S)
Date(s) 05/02/2021 - 05/02/2021
linked URL https://indico.math.cnrs.fr/event/6351/

All the collection videos (6)

49:26

published on February 5, 2021

Supervised Learning with Missing Values

By Gaël Varoquaux

38:55

published on February 5, 2021

Analysis of Gradient Descent on Wide Two-Layer Neural Networks

By Lenaïc Chizat

45:29

published on February 5, 2021

Deep Unfolding of a Proximal Interior Point Method for Image Restoration

By Emilie Chouzenoux

42:00

published on February 5, 2021

Maximum Entropy Distributions for Image Synthesis under Statistical Constraints

By Agnès Desolneux

41:20

published on February 5, 2021

Input Similarity from the Neural Network Perspective

By Guillaume Charpiat

43:37

published on February 5, 2021

Deep Neural Network for Audio and Music Transformations

By Gaël Richard

Mathematics
and Interactions

Complex foliations and differential equations / Feuilletages et équations différentielles complexes

Outreach

CEMRACS 2023: Scientific Machine Learning / CEMRACS 2023: Apprentissage automatique scientifique

Adaptive and High-Dimensional Spatio-Temporal Methods for Forecasting / Méthodes spatio-temporelles adaptatives et en grande dimension pour la prédiction

Collection Statistics & Computer Science Day for Data Science at Paris-Saclay 2021

Supervised Learning with Missing Values

Analysis of Gradient Descent on Wide Two-Layer Neural Networks

Deep Unfolding of a Proximal Interior Point Method for Image Restoration

Maximum Entropy Distributions for Image Synthesis under Statistical Constraints

Input Similarity from the Neural Network Perspective

Deep Neural Network for Audio and Music Transformations

Score matching for simulating sub-Riemannian diffusion bridge processes

High-dimentional classification with deep neural networks: decision boundaries, noise, and margin

Barycenters for transport costs