University of Cambridge > Talks.cam > Isaac Newton Institute Seminar Series > Rate estimates for total variation distance

  Browse A-Z

Log in

University Account External (via Google)

🔐 Log In

Information on

Subscribing to talksdetails Finding a talkdetails Adding a talkdetails Disseminating talksdetails Help and Documentationdetails

ℹī¸ Information

Rate estimates for total variation distance

Download to your calendar using vCal

  • UserMiklos Rasonyi (HUN-REN Renyi Institute)
  • ClockFriday 19 July 2024, 16:00-17:00
  • HouseExternal.

If you have a question about this talk, please contact nobody.

DMLW01 - International workshop on diffusions in machine learning: foundations, generative models, and optimisation

For algorithms of machine learning, rate estimates are often provided in the Wasserstein metric or some variant thereof. There has been spectacular recent progress in the techniques for establishing such estimates.At the same time, powerful methods have been developed in Malliavin calculus that enable to infer total variation convergence from weak convergence.After presenting an overview of the developments above, we show some new results on total variation convergence that do not rely on Malliavin calculus nevertheless they are applicable to various discrete-time numerical schemes. This talk is based on joint work with I. Ivkovic.  

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

Note that ex-directory lists are not shown.

Other lists

Gonville & Caius College, Cambridge: the Green Room JDB Seminar Room, CUED Cambridge Enterprise

Other talks

Laser tomography for wide area gas emission spatio-temporal reconstruction Cambridge Reading Group on Reproduction: Professor Sara Brill An AI-Assisted Proof in Spectral Geometry Cambridge RNA Club - ONLINE BSU Seminar: "Response Adaptivity Across Clinical Trials In Portfolios of Biomedical Innovations" A Few Days in Patagonia - the Deep South
 

Š 2006-2025 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity