Efficient Approximate Bayesian Inference (25w5335)


Sean Plummer (University of Arkansas)

Debdeep Pati (Texas A&M University)

Dootika Vats (Indian Institute of Technology Kanpur)

Yun Yang (University of Illinois Urbana-Champaign)

Shuang Zhou (Arizona State University)


The Banff International Research Station will host the “Efficient Approximate Bayesian Inference” workshop in Banff from March 9 to 14, 2025.

Variational inference (VI) is an optimization based approach to approximate inference, originating from ideas in statistical physics, which has rapidly grown in popularity over the last two decades due to its ability to achieve an impressive level of computational scalability while maintaining a reasonable level of statistical accuracy. VI provides statisticians with a powerful tool capable of analyzing the large and complex datasets arising from modern scientific investigations and studies in the social sciences. Despite the large body of methodological work aimed at developing increasing powerful approaches to VI, the statistical properties of these approximations are known only in the simplest case, known as mean-field VI.

This workshop aims to bring together a diverse and multi-disciplinary group of computer scientists, statisticians, and physicists to bridge ideas and create working groups aimed toward (1) further improving variational methodology for models of complex datasets, (2) advancing the theoretical understanding for variational inference beyond the mean-field setting, and (3) developing deeper interdiciplanary connections between physics, computer science, and statistics. Results from the workshop will ultimately help to advance the use of these recently developed, flexible approaches to variational inference which have been developed for models of complex data arising in scientific applications.

The Banff International Research Station for Mathematical Innovation and Discovery (BIRS) is a collaborative Canada-US-Mexico venture that provides 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 is located at The Banff Centre in Alberta and is supported by Canada’s Natural Science and Engineering Research Council (NSERC), the U.S. National Science Foundation (NSF), and Alberta’s Advanced Education and Technology.