Optimal Transport + Machine Learning (OT + ML)

Zaid Harchaoui (teaching) (University of Washington) , Soumik Pal (teaching) (University of Washington) , Young-Heon Kim (WDA administrator) (University of British Columbia)

Sep 1, 2021 — Dec 31, 2021

About the course

In the second installment of OT+X series we take X=ML or machine learning. A number of problems equivalent or related to the Monge-Kantorovich Optimal Transport (OT) problem have appeared in recent years in machine learning, and data science at large. The fruitful connections between the two fields have led to several important advances impacting both. The Wasserstein metric defines a metric between probability measures, used to describe distributions over data or distributions over models, that improves upon existing metrics based on Hilbertian metrics and f-divergences, and that is now more easily amenable to efficient numerical computation.

The first part of the course will cover the mathematical basics of OT and introduce the geometry of Wasserstein spaces. The second part of the course will cover computational aspects of OT and describe the central role played by OT in convergence analysis of stochastic algorithms for deep learning, in distributionally robust statistical learning, and in combinatorial or geometrical problems arising in data science applications. The course is meant for a wide audience including graduate students and industry professionals. Prior knowledge of real analysis, probability, statistics, and machine learning will be particularly helpful. The course will be interspersed with numerical illustrations. Familiarity with coding in Python or R is a plus.

Registration

This course is available for registration under the Western Dean's Agreement. To register, you must obtain the approval of the course instructor and you must complete the Western Dean's agreement form , using the details below. The completed form should be signed by your home institution department and school of graduate studies, then returned to the host institution of the course.

Enrollment Details

Course Name
Optimal Transport + Machine Learning (OT + ML)
Date
Sep 1, 2021 — Dec 31, 2021
Course Number
Math 566 - Topics in Optimal Transport
Section Number
Section Code

Instructor(s)

  • Zaid Harchaoui (teaching)

    University of Washington

  • Soumik Pal (teaching)

    University of Washington

  • Young-Heon Kim (WDA administrator)

    University of British Columbia

For help with completing the Western Dean’s agreement form, please contact the graduate student program coordinator at your institution. For more information about the agreement, please see the Western Dean's Agreement website

Other Course Details

2021-2022