This event is part of Decision Making and Uncertainty View Details

Applied Optimal Transport

Description

Back to top

This workshop showcases current developments in theoretical and computational optimal transport with a focus on applications in machine learning and statistics.

Organizers

Back to top
A G
Aude Genevay MIT
M N
Marcel Nutz Columbia University
B S
Bodhisattva Sen Columbia University

Speakers

Back to top
D A
David Alvarez-Melis Microsoft Research
F B
Francis Bach INRIA
J B
Julio Backhoff-Veraguas Universitaet Wien
J B
Jose Blanchet Stanford University
G C
Guillaume Carlier Université Paris Dauphine
E C
Elsa Cazelles Centre National de la Recherche Scientifique (CNRS)
L C
Lenaic Chizat Centre National de la Recherche Scientifique (CNRS)
M C
Marco Cuturi Apple ML Research and École Nationale de la Statistique et de l’Administration Économique
N D
Nabarun Deb Columbia University
E d B
Eustasio del Barrio Universidad de Valladolid
A G
Alfred Galichon New York University
P G
Promit Ghosal MIT
F G
Florian Gunsilius University of Michigan
J K
Jun Kitagawa Michigan State University
A K
Anna Korba École Nationale de la Statistique et de l’Administration Économique
G M
Gonzalo Mena University of Oxford
L N
Long Nguyen University of Michigan
J N
Jonathan Niles-Weed New York University
S P
Soumik Pal University of Washington
J S
Johan Segers UCLouvain
J W
Johannes Wiesel Columbia University

Schedule

Back to top
Monday, May 16, 2022
9:30-10:15 CDT
Gradient flows on graphons

Speaker: Soumik Pal (University of Washington)

10:30-11:15 CDT
Quantitative geometric stability for semi-discrete optimal transport

Speaker: Jun Kitagawa (Michigan State University)

11:30-13:00 CDT
Lunch
13:00-13:45 CDT
Inverse bounds for learning latent structures

Speaker: Long Nguyen (University of Michigan)

14:00-14:45 CDT
Machine Learning in the Space of Datasets: an Optimal Transport Perspective

Speaker: David Alvarez-Melis (Microsoft Research & Harvard University)

15:00-15:30 CDT
Break
15:30-16:15 CDT
The Wasserstein-Martingale projection of a Brownian motion given initial and terminal marginals.

Speaker: Julio Backhoff-Veraguas (Universitaet Wien)

Tuesday, May 17, 2022
9:30-10:15 CDT
Information theory with kernel methods

Speaker: Francis Bach (Institut National de Recherche en Informatique et Automatique (INRIA))

10:30-11:15 CDT
Wasserstein entropic barycenters

Speaker: Guillaume Carlier (Université Paris Dauphine)

11:30-13:00 CDT
Lunch
13:00-13:45 CDT
Distributionally Robust Gaussian Process Regression

Speaker: Jose Blanchet (Stanford University)

14:00-14:45 CDT
Towards practical estimation of Brenier maps

Speaker: Jonathan Niles-Weed (Courant Institute of Mathematical Sciences)

15:00-15:30 CDT
Break
15:30-16:15 CDT
Measuring association with Wasserstein distances

Speaker: Johannes Wiesel (Columbia University)

Wednesday, May 18, 2022
9:30-10:15 CDT
Differentiable Matchings, Mappings and JKO Steps

Speaker: Marco Cuturi (Apple ML Research and École Nationale de la Statistique et de l’Administration Économique)

10:30-11:15 CDT
A novel notion of barycenter for probability distributions based on optimal weak mass transport

Speaker: Elsa Cazelles (Centre National de la Recherche Scientifique (CNRS))

11:30-13:00 CDT
Lunch
13:00-13:45 CDT
Effect of Dependence on the Convergence of Empirical Wasserstein Distance

Speaker: Nabarun Deb (Columbia University)

14:00-14:45 CDT
On model-based clustering with entropic optimal transport

Speaker: Gonzalo Mena (University of Oxford)

15:00-15:30 CDT
Break
15:30-16:15 CDT
Geometry and Stability of Entropic Optimal Transport

Speaker: Promit Ghosal (Massachusetts Institute of Technology (MIT))

Thursday, May 19, 2022
9:30-10:15 CDT
The regularized equilibrium transport problem: analysis, computation and economic applications

Speaker: Alfred Galichon (New York University)

10:30-11:15 CDT
Uniform consistency of estimated optimal transport plans

Speaker: Johan Segers (UCLouvain)

11:30-13:00 CDT
Lunch
13:00-13:45 CDT
Matching for causal effects via multimarginal optimal transport

Speaker: Florian Gunsilius (University of Michigan)

14:00-14:45 CDT
Trajectory Inference via Mean-Field Langevin in Path Space

Speaker: Lenaic Chizat (EPFL)

15:00-16:00 CDT
Social Hour
Friday, May 20, 2022
9:30-10:15 CDT
Sampling with kernelized Wasserstein gradient flows

Speaker: Anna Korba (ENSAE)

10:30-11:15 CDT
Nonparametric Multiple-Output Center-Outward Quantile Regression

Speaker: Eustasio del Barrio (Universidad de Valladolid)


Videos

Back to top

Gradient flows on graphons

Soumik Pal
May 16, 2022

Quantitative geometric stability for semi-discrete optimal transport

Jun Kitagawa
May 16, 2022

Inverse bounds for learning latent structures

Long Nguyen
May 16, 2022

The Wasserstein-Martingale projection of a Brownian motion given initial and terminal marginals.

Julio Backhoff-Veraguas
May 16, 2022

Distributionally Robust Gaussian Process Regression

Jose Blanchet
May 17, 2022

Towards practical estimation of Brenier maps

Jonathan Niles-Weed
May 17, 2022

Differentiable Matchings, Mappings and JKO Steps

Marco Cuturi
May 18, 2022

Effect of Dependence on the Convergence of Empirical Wasserstein Distance

Nabarun Deb
May 18, 2022

Geometry and Stability of Entropic Optimal Transport

Promit Ghosal
May 18, 2022

Uniform consistency of estimated optimal transport plans

Johan Segers
May 19, 2022

Trajectory Inference via Mean-Field Langevin in Path Space

Lenaic Chizat
May 19, 2022

Sampling with kernelized Wasserstein gradient flows

Anna Korba
May 20, 2022

Nonparametric Multiple-Output Center-Outward Quantile Regression

Eustasio del Barrio
May 20, 2022