MRC Biostatistics Unit 23rd Armitage Keynote Lecture

Speaker: Professor David M. Blei, Professor of Statistics and Computer Science from Columbia University
Series: MRC Biostatistics Armitage Workshops and Lectures
Note: Registration: https://onlinesales.admin.cam.ac.uk/conferences-and-events/mrc-biostatistics-unit/armitage-lecture-workshop/23rd-armitage-workshop-and-lecture-10-november-2026

Empirical Bayes improves simultaneous inference by learning from related data. In this talk, I will present three recent directions in empirical Bayes. First, I will discuss an empirical Bayes approach to combining randomized experiments and observational studies, where calibration studies make it possible to learn the distribution of observational bias and use observational data in a principled way. Second, I will discuss empirical Bayes for implicit likelihoods, where the model is available only through a simulator, and show how simulation-based inference can be used to produce empirical Bayes estimates without evaluating a density. Third, I will discuss a general method based on probabilistic symmetries, which extends empirical Bayes beyond exchangeable settings to structured problems such as arrays, graphs, conditional data, and spatial models. These three ideas illustrate new roles for empirical Bayes in modern statistics and machine learning.

This is joint work with Diana Cai, Don Green, Sebastian Salazar, Xinwei Shen, Eli Weinstein, Bohan Wu, Cheng Zhang.

This will be the keynote lecture in a full day of talks on the theme of Bayesian Machine Learning.