Updated with Zoom link: Complexity of Optimal Transport

Abstract

Quals talk:

Optimal transport (OT) is a tool for measuring distances between probability distributions. Recently OT has found use in GANs, bounds on generalization errors in statistical machine learning, domain adaptation, and domain generalization, to name a few. In all of these contexts, an important problem is to estimate the OT distance from samples. The use of finite samples brings many important questions concerning both the statistical efficiency and the computational complexity in estimating OT maps under various models of observing the underlying distributions. In this talk, I will discuss a collection of recent results on these problems as well as discuss connections to ongoing works at Tufts.

Please join meeting in Halligan 209 or via Zoom.

Join Zoom Meeting: https://tufts.zoom.us/j/97183120811

Meeting ID: 971 8312 0811

Password: See colloquium email

Dial by your location: +1 646 558 8656 US (New York)

Meeting ID: 971 8312 0811

Passcode: See colloquium email