The fifty-seventh UQSay seminar on UQ, DACE and related topics will take place online on Thursday afternoon, March 30, 2023.
WARNING: Unusual starting time: UQSay #57 will begin at 3 PM (Paris time). Don’t trust the poster.
Bayesian inference makes more sense for modern neural networks than virtually every other model class, because these models can represent many compelling and complementary explanations for data, corresponding to different settings of their parameters. However, a number of myths have emerged about modern Bayesian deep learning. In this talk we will evaluate the following questions: (1) is Bayesian deep learning practical? (2) are standard (e.g. Gaussian) priors arbitrary and poor? (3) is “deep ensembles” a non-Bayesian competitor to standard approximate inference approaches? (4) does the common practice of posterior tempering, leading to “cold posteriors”, mean the Bayesian posterior is poor? (5) is the marginal likelihood a reasonable way to select between trained networks? I will also discuss the success stories, future opportunities, and challenges in Bayesian deep learning.
Organizing committee: Pierre Barbillon (MIA-Paris), Julien Bect (L2S), Nicolas Bousquet (EDF R&D), Amélie Fau (LMPS), Filippo Gatti (LMPS), Bertrand Iooss (EDF R&D), Alexandre Janon (LMO), Sidonie Lefebvre (ONERA), Didier Lucor (LISN), Emmanuel Vazquez (L2S).
Coordinators: Julien Bect (L2S) & Sidonie Lefebvre (ONERA)
Practical details: the seminar will be held online using Microsoft Teams.
If you want to attend this seminar (or any of the forthcoming online UQSay seminars), and if you do not already have access to the UQSay group on Teams, simply send an email and you will be invited. Please specify which email address the invitation must be sent to (this has to be the address associated with your Teams account).
You will find the link to the seminar on the “General” UQSay channel on Teams, approximately 15 minutes before the beginning.
The technical side of things: you can use Teams either directly from your web browser or using the “fat client”, which is available for most platforms (Windows, Linux, Mac, Android & iOS). We strongly recommend the latter option whenever possible. Please give it a try before the seminar to anticipate potential problems.