Lectures on « High-Dimensional Data Analysis » (GDR MIA) – INSTITUT DE MATHÉMATIQUES DE MARSEILLE

Journées
FRUMAM, St Charles, Marseille
https://www.i2m.univ-amu.fr/seminaires_signal_apprentissage/Conf/Oct2021/index.php

Date(s) : 21/10/2021 - 22/10/2021 iCal
0 h 00 min

In the frame of the GdR MIA

The analysis of high-dimensional data sets is now ubiquitous in many applications and blends tools from probability theory, machine learning, geometry, graph theory, statistics and optimization. The objective of this two-day event is to gather researchers from different backgrounds to explore recent advances in this field and to stimulate discussions. Special attention will be given to young researchers who are encouraged to participate with an introductory lecture on the first day and dedicated funding for their travel and accommodation.

Confirmed speakers

Elsa Cazelles (CNRS IRIT, Toulouse)
Lenaïc Chizat (CNRS, Orsay)
Agnès Desolneux (CNRS Centre Borelli, Paris)
Aurélien Garivier (ENS Lyon)
Nicolas Keriven (CNRS GIPSA, Grenoble)
Olga Mula (CEREMADE, Paris)
Geneviève Robin (LaMME, ́Evry)
François-Xavier Vialard (LIGM, Marne-la-Vallée).

Program

		Thursday, October, 21th
8:45		Welcome
9.00-17.30		Measure concentration and statistics in high-dimension: an introduction This course is dedicated to students and any person willing to acquire a background in measure concentration.
		Friday, November, 15th
9:30-10:00		Coffee and introduction
10.00-10.45	Agnès Desolneux	A Wasserstein-type distance in the space of Gaussian mixture models Abstract.

10.45-11:15	François-Xavier Vialard	TBD Abstract.

11:15-11:45		Break
11:45-12:15	Elsa Cazelles	A novel notion of barycenter for probability distributions based on optimal weak mass transport Abstract.

12:15-14:00		Lunch
14:00-14:45	Lénaïc Chizat	Analysis of gradient descent on wide two-layer ReLU neural networks Abstract.

14:45-15:15	Geneviève Robin	Outliers detection in networks with missing links Abstract.

15:15-15:45		Break
15:45-16:15	Nicolas Keriven	Graph neural networks on large random graphs: convergence, stability, universality Abstract.

16:15-16:45	Olga Mula	Model order reduction by interpolation in Wasserstein spaces Abstract.

16:45-17:00		Conclusions

Organizers

Paul Escande.
Caroline Chaux.
Clothilde Mélot.

Emplacement
FRUMAM, St Charles (2ème étage)

Catégories