DROPS

Document

Rethinking the Role of Bayesianism in the Age of Modern AI (Dagstuhl Seminar 24461)

Authors: Vincent Fortuin, Mohammad Emtiyaz Khan, Mark van der Wilk, Zoubin Ghahramani, and Katharine Fisher

Published in: Dagstuhl Reports, Volume 14, Issue 11 (2025)

Abstract

Despite the recent success of large-scale deep learning, these systems still fall short in terms of their reliability and trustworthiness. They often lack the ability to estimate their own uncertainty in a calibrated way, encode meaningful prior knowledge, avoid catastrophic failures, and also reason about their environments to avoid such failures. Since its inception, Bayesian deep learning (BDL) has harbored the promise of achieving these desiderata by combining the solid statistical foundations of Bayesian inference with the practically successful engineering solutions of deep learning methods. This was intended to provide a principled mechanism to add the benefits of Bayesian learning to the framework of deep neural networks. However, compared to its promise, BDL methods often do not live up to the expectation and underdeliver in terms of real-world impact. This is due to many fundamental challenges related to, for instance, computation of approximate posteriors, unavailability of flexible priors, but also lack of appropriate testbeds and benchmarks. To make things worse, there are also numerous misconceptions about the scope of Bayesian methods, and researchers often end up expecting more than what they can get out of Bayes. By bringing together researchers from diverse communities, such as machine learning, statistics, and deep learning practice, in a personal and interactive seminar environment featuring debates, round tables, and brainstorming sessions, our Dagstuhl Seminar "Rethinking the Role of Bayesianism in the Age of Modern AI" (24461) has discussed these questions from a variety of angles and charted a path for future research to innovate, enhance, and strengthen meaningful real-world impact of Bayesian deep learning.

Cite as

Vincent Fortuin, Mohammad Emtiyaz Khan, Mark van der Wilk, Zoubin Ghahramani, and Katharine Fisher. Rethinking the Role of Bayesianism in the Age of Modern AI (Dagstuhl Seminar 24461). In Dagstuhl Reports, Volume 14, Issue 11, pp. 40-59, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2025)

Copy BibTex To Clipboard

@Article{fortuin_et_al:DagRep.14.11.40,
  author =	{Fortuin, Vincent and Khan, Mohammad Emtiyaz and van der Wilk, Mark and Ghahramani, Zoubin and Fisher, Katharine},
  title =	{{Rethinking the Role of Bayesianism in the Age of Modern AI (Dagstuhl Seminar 24461)}},
  pages =	{40--59},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2025},
  volume =	{14},
  number =	{11},
  editor =	{Fortuin, Vincent and Khan, Mohammad Emtiyaz and van der Wilk, Mark and Ghahramani, Zoubin and Fisher, Katharine},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.14.11.40},
  URN =		{urn:nbn:de:0030-drops-228200},
  doi =		{10.4230/DagRep.14.11.40},
  annote =	{Keywords: Bayesian machine learning, deep learning, foundation models, model selection, uncertainty estimation}
}

Document

DOI: 10.4230/DagRep.13.2.47

Challenges and Perspectives in Deep Generative Modeling (Dagstuhl Seminar 23072)

Authors: Vincent Fortuin, Yingzhen Li, Kevin Murphy, Stephan Mandt, and Laura Manduchi

Published in: Dagstuhl Reports, Volume 13, Issue 2 (2023)

Abstract

Deep generative models, such as variational autoencoders, generative adversarial networks, normalizing flows, and diffusion probabilistic models, have attracted a lot of recent interest. However, we believe that several challenges hinder their more widespread adoption: (C1) the difficulty of objectively evaluating the generated data; (C2) challenges in designing scalable architectures for fast likelihood evaluation or sampling; and (C3) challenges related to finding reproducible, interpretable, and semantically meaningful latent representations. In this Dagstuhl Seminar, we have discussed these open problems in the context of real-world applications of deep generative models, including (A1) generative modeling of scientific data, (A2) neural data compression, and (A3) out-of-distribution detection. By discussing challenges C1-C3 in concrete contexts A1-A3, we have worked towards identifying commonly occurring problems and ways towards overcoming them. We thus foresee many future research collaborations to arise from this seminar and for the discussed ideas to form the foundation for fruitful avenues of future research. We proceed in this report by summarizing the main results of the seminar and then giving an overview of the different contributed talks and working group discussions.

Cite as

Vincent Fortuin, Yingzhen Li, Kevin Murphy, Stephan Mandt, and Laura Manduchi. Challenges and Perspectives in Deep Generative Modeling (Dagstuhl Seminar 23072). In Dagstuhl Reports, Volume 13, Issue 2, pp. 47-70, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)

Copy BibTex To Clipboard

@Article{fortuin_et_al:DagRep.13.2.47,
  author =	{Fortuin, Vincent and Li, Yingzhen and Murphy, Kevin and Mandt, Stephan and Manduchi, Laura},
  title =	{{Challenges and Perspectives in Deep Generative Modeling (Dagstuhl Seminar 23072)}},
  pages =	{47--70},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2023},
  volume =	{13},
  number =	{2},
  editor =	{Fortuin, Vincent and Li, Yingzhen and Murphy, Kevin and Mandt, Stephan and Manduchi, Laura},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.13.2.47},
  URN =		{urn:nbn:de:0030-drops-191817},
  doi =		{10.4230/DagRep.13.2.47},
  annote =	{Keywords: deep generative models, representation learning, generative modeling, neural data compression, out-of-distribution detection}
}

Search Results

Documents authored by Fortuin, Vincent

Rethinking the Role of Bayesianism in the Age of Modern AI (Dagstuhl Seminar 24461)

Abstract

Cite as

Challenges and Perspectives in Deep Generative Modeling (Dagstuhl Seminar 23072)

Abstract

Cite as

Thanks for your feedback!

Could not send message