Bayesian nonparametrics combines the flexibility often associated with machine learning with principled uncertainty quantification required for inference. Popular priors in this class include Gaussian Processes, Bayesian Additive Regression Trees, Chinese Restaurant Processes, and more. But what exactly are “nonparametric” priors? How can we compute posteriors under such priors? And how can we use them for flexible modeling? This talk will explore these questions by introducing nonparametric Bayes at a conceptual level and walking through a few common priors, with a particular focus on the Dirichlet Process prior for regression.
Sorry about the video problems at the beginning. We are working to sort that out in future videos.
Arman is an Assistant Professor of Biostaistics (tenure-track) at Brown University’s Department of Biostatistics. His methodological research centers around developing nonparametric Bayesian methods for flexibly estimating causal effects with observational data. This involves, among other things, devising priors over high-dimensional model spaces and constructing MCMC methods for efficient posterior computation. Though broadly applicable, these methodological interests are motivated by applied problems in health economics and, more recently, statistical challenges in oncology research.
Arman's website: https://stablemarkets.netlify.app/