Bayesian Uncertainty Quantification in Large Models (25w5329)

Organizers

Subhashis Ghoshal (North Carolina State University)

Marta Catalano (Luiss University)

Fengnan Gao (University College Dublin)

Igor Pruenster (Bocconi University, Italy)

Yanxun Xu (Johns Hopkins University)

Description

The Chennai Mathematical Institute will host the "Bayesian Uncertainty Quantification in Large Models" workshop in Chennai, India from December 7 to December 12, 2025.


Making decisions in the face of uncertainty is a fundamental aspect of life, from personal actions to business decisions to government policies. For scientific decision-making, it is essential that the uncertainty in such decision-making is quantified by numerical assessments. In statistical and machine learning analysis, optimal decisions are made based on available information. The Bayesian approach addresses the issue by updating information about unknown parameters quantified as a random probability distribution, starting with an initial assessment of the parameters described by a prior probability distribution. The fundamental principle behind the Bayesian approach is logically straightforward. However, challenges remain in proposing appropriate prior distributions, devising methods of posterior updating, and theoretically studying the quality of Bayesian uncertainty quantification. In the workshop, a group of leading experts and emerging researchers will meet to discuss various aspects of Bayesian uncertainty quantification in different situations, challenges, and solutions. The workshop will give Asian researchers an exceptional opportunity to participate, and especially emerging researchers will get an excellent exposure to this vast topic.


The Chennai Mathematical Institute (CMI) in Chennai, India, and the Banff International Research Station for Mathematical Innovation and Discovery (BIRS) in Banff, are collaborative Canada-US-Mexico ventures that provide an environment for creative interaction as well as the exchange of ideas, knowledge, and methods within the Mathematical Sciences, with related disciplines and with industry. The research station in Banff is supported by Canada's Natural Science and Engineering Research Council (NSERC), the U.S. National Science Foundation (NSF), and Alberta's Advanced Education