Measure Transport, Diffusion Processes and Sampling Workshop

Date & Time

This workshop aims to foster discussions in machine learning underlying modern generative models and their role in MCMC-like sampling algorithms as well as explore how viewpoints arising in adjacent fields can be used to refine and approach remaining open challenges in the contexts of probabilistic modeling and high-dimensional sampling.

The past decade has seen a surge of progress in the empirical performance of techniques such as diffusion models and normalizing flows, but much of the success of these techniques amounts to careful consideration of how to define a map between distributions.

This topic has a substantial history in the fields of optimal transport, stochastic processes, and variational inference. By bringing together theorists and practitioners from these camps, we hope to clarify the perspectives of recent advances across their respective communities.

Workshop Agenda

TimeSpeakerTalk Title
Monday, December 4th
8:00 a.m. - 9:00 a.m. Breakfast
9:00 a.m. - 10:00 a.m.Session 1 Gareth Roberts, University of WarwickBayesian Fusion
10:00 a.m. - 11:00 a.m.Session 2Zahra Kadkhodaie, New York UniversityGeneralization in diffusion models arises from Geometry-adaptive harmonic representation
11:00 a.m. - 11:30 a.m. Coffee BreakCoffee Break
11:30 a.m. - 12:30 p.m. Spotlight AHolden Lee, Johns Hopkins
Lorenz Richter, Zuse Institute Berlin
Semon Rezchikov, Princeton University
12:30 p.m. - 3:00 p.m.Lunch & Collaboration
3:00 p.m.Session 3Katy Craig, University of California, Santa BarbaraNonlocal Approximation of Fast and Slow Diffusion
4:00 p.m. - 5:00 p.m.Session 4Jianfeng Lu, Duke University
Tuesday, December 5th
8:00 a.m. - 9:00 a.m.Breakfast
9:00 a.m. - 10:00 a.m.Session 5Eric Vanden-Eijnden, Courant Institute, NYUStochastic Interpolants: A unifying framework for flows and diffusions
10:00 a.m. - 11:00 a.m.Session 6Andrea Montanari, Stanford UniversitySampling from Gibbs measures and Bayes posteriors via diffusion processes
11:00 a.m. - 11:30 a.m.Coffee Break
11:30 a.m. - 12:30 p.m. Spotlight BBruno Régaldo-Saint Blancard, Flatiron Institute
Ahmed El Alaoui, Cornell University
Gabriele Corso, MIT
12:30 p.m. - 3:00 p.m.Lunch & Working Together
3:00 p.m. - 4:00 p.m.Session 7Andre Wibisono, Yale UniversitySampling with Estimated Score Functions
4:00 p.m. - 6:00 p.m.Poster Session & Reception
Wednesday, December 6th
8:00 a.m. - 9:00 a.m.Breakfast
9:00 a.m. - 10:00 a.m.Session 8Jonathan Niles-Weed, New York UniversityOptimal transport map estimation in general function spaces
10:00 a.m. - 11:00 a.m.Session 9Ricky Chen, MetaGenerative Flows: Applications Beyond Distribution Matching
11:00 a.m. - 11:30 a.m.Coffee Break
11:30 a.m. - 12:30 p.m.Spotlight CUroš Seljiak, University of California Berkeley
Yuansi Chen, Duke University
Valentin de Bortoli, Google Deep Mind
12:30 p.m. - 3:00 p.m.Lunch & Working Together
3:00 p.m. - 4:00 p.m.Session 10Qin Li, University of Wisconsin-MadisonAccelerating optimization over probability measure space
4:00 p.m. - 5:00 p.m.Session 11Accelerating optimization over probability measure spaceDiffusion-based variational inference
Advancing Research in Basic Science and MathematicsSubscribe to Flatiron Institute announcements and other foundation updates