Workflow Workshop - Dr. Debdeep Pati

• Time: Monday 10/16/2023 and 10/23/2023 from 10:30 AM to 12:30 PM
• Location: BLOC 503

Topic

Introduction to Bayesian Asymptotics

Description

In the first part of the session, Dr. Pati plans to go over the general setup to study the asymptotic behavior of the posterior and motivate the sufficient conditions needed for the posterior to concentrate around the true parameter. There will be emphasis on the three key requirements – how to ensure that the prior is adequately concentrated around the true parameter, how one can distinguish between two parameters in the model space using test functions (identifiability), with reference to the works of LeCam-Birgé, and how to analyze the complexity of the parameter space.

The workshop is intended for students and researchers who are either pursuing theoretical research involving Bayesian statistics or are interested in providing theoretical support for methodological developments. Background on probability theory and real analysis (at the PhD level) is required.

Additional Resources

• Dr. Pati’s Webpage
• Fundamentals of Nonparametric Bayesian Inference
• Notes by Bas Kleijn, Aad van der Vaart, Harry van Zanten

First Session Recording

Second Session Recording

Gallery