Logo

Mathematical Sciences Research Institute

Home » [Moved Online] Hot Topics: Optimal transport and applications to machine learning and statistics

Workshop

[Moved Online] Hot Topics: Optimal transport and applications to machine learning and statistics May 04, 2020 - May 08, 2020
Registration Deadline: May 08, 2020 3 months ago
To apply for Funding you must register by: February 04, 2020 6 months ago
Parent Program: --
Series: Hot Topic, Hot Topic
Location: MSRI: Simons Auditorium, Atrium
Organizers Luigi Ambrosio (Scuola Normale Superiore), Francis Bach (École Normale Supérieure; Institut National de Recherche en Informatique Automatique (INRIA)), LEAD Katy Craig (University of California, Santa Barbara), Carola-Bibiane Schönlieb (University of Cambridge), Stefano Soatto (University of California, Los Angeles)
Speaker(s)

Show List of Speakers

Description
Image
Image drawn by Dr. Katy Craig
This workshop will be held online.  The link to join is: https://msri.zoom.us/j/92457794010. You must register for the workshop to receive the password.  The workshop is held in Pacific Standard Time. Workshop Description: The goal of the workshop is to explore the many emerging connections between the theory of Optimal Transport and models and algorithms currently used in the Machine Learning community. In particular, the use of Wasserstein metrics and the relation between discrete models and their continuous counterparts will be presented and discussed. Bibliography
Schedule, Notes/Handouts & Videos
Show Schedule, Notes/Handouts & Videos
Show All Collapse
May 04, 2020
Monday
09:15 AM - 09:30 AM
  Welcome
David Eisenbud (MSRI - Mathematical Sciences Research Institute)
09:30 AM - 10:30 AM
  Scaling Optimal Transport for High dimensional Learning
Gabriel Peyré (École Normale Supérieure)
10:30 AM - 11:00 AM
  Break
11:00 AM - 12:00 PM
  Linear Unbalanced Optimal Transport
Matthew Thorpe (University of Manchester)
12:00 PM - 02:00 PM
  Break
02:00 PM - 03:00 PM
  Computing Wasserstein barycenters using gradient descent algorithms
Philippe Rigollet (Massachusetts Institute of Technology)
May 05, 2020
Tuesday
09:30 AM - 10:30 AM
  Kalman-Wasserstein Gradient Flows
Franca Hoffman (California Institute of Technology)
10:30 AM - 11:00 AM
  Break
11:00 AM - 12:00 PM
  A Deeper Understanding of the Quadratic Wasserstein Metric in Inverse Data Matching
Yunan Yang (New York University, Courant Institute)
12:00 PM - 02:00 PM
  Break
02:00 PM - 03:00 PM
  A Machine Learning Framework for Solving High-Dimensional Mean Field Game and Mean Field Control Problems
Samy Wu Fung (University of California, Los Angeles)
May 06, 2020
Wednesday
09:30 AM - 10:30 AM
  From quantization of measures to weighted ultrafast diffusion equations
Mikaela Iacobelli (ETH Zurich)
10:30 AM - 11:00 AM
  Break
11:00 AM - 12:00 PM
  Equality of the Jellium and Uniform Electron Gas next-order asymptotic terms for Coulomb and Riesz potentials
Codina Cotar (University College)
May 07, 2020
Thursday
09:30 AM - 10:30 AM
  Regularity theory and uniform convergence in the large data limit of graph Laplacian eigenvectors on random data clouds.
Nicolas Garcia Trillos (University of Wisconsin-Madison)
10:30 AM - 11:00 AM
  Break
11:00 AM - 12:00 PM
  From an ODE to accelerated stochastic gradient descent: convergence rate and empirical results
Adam Oberman
12:00 PM - 02:00 PM
  Break
02:00 PM - 03:00 PM
  Learning with Few Labeled Data
Pratik Chaudhari (University of Pennsylvania)
03:00 PM - 03:30 PM
  Break
03:30 PM - 04:30 PM
  Fusion with Optimal Transport
Justin Solomon (Massachusetts Institute of Technology)
May 08, 2020
Friday
09:30 AM - 10:30 AM
  Learning with entropy-regularized optimal transport
Aude Genevay (Massachusetts Institute of Technology)
10:30 AM - 11:00 AM
  Break
11:00 AM - 12:00 PM
  Analysis of Gradient Descent on Wide Two-Layer ReLU Neural Networks
Lenaic Chizat (Centre National de la Recherche Scientifique (CNRS))
12:00 PM - 02:00 PM
  Break
02:00 PM - 03:00 PM
  Mean field theory of neural networks: From stochastic gradient descent to Wasserstein gradient flows
Andrea Montanari (Stanford University)
03:00 PM - 03:30 PM
  Break
03:30 PM - 04:30 PM
  Nonlocal-interaction equations on graphs and gradient flows in nonlocal Wasserstein metric
Dejan Slepcev (Carnegie Mellon University)