Stat Cafe - Dr. Yu-Chien Bo Ning
Recent Progress and Challenges in Bridging Bayesian Survival Analysis and Deep Learning: Theory, Computation, and Applications
- Time: Tuesday, 10/06/2026 from 11:10AM to 12:25PM
- Location: BLOC 448
Description
Despite rapid advances in Bayesian deep learning, progress in developing Bayesian deep learning methods for survival analysis has been relatively slow. In this talk, I will discuss the theoretical, computational, and practical obstacles to developing Bayesian deep learning methods for survival outcomes. I will begin with the rich history of Bayesian survival analysis in low-dimensional settings and then present my recent work on extending these methods to high-dimensional and nonparametric models. I will also discuss state-of-the-art computational approaches and the modeling challenges posed by large-scale electronic health record (EHR) studies. Throughout the talk, I will highlight how these developments connect to deep learning and what remains to be done to build methods that are theoretically sound, computationally scalable, and useful in practice.
Our Speaker
Dr. Yu-Chien Bo Ning is an assistant professor in the Department of Statistics at Texas A&M University. His research focuses on developing trustworthy Bayesian machine learning methods for data science. Previously, he held positions at Harvard HSPH (Research Associate), UC Davis (Visiting AP), Sorbonne Université in Paris (FSMP Postdoctoral Fellowship), and Yale (Postdoc). He obtained a Ph.D. in Statistics at North Carolina State University.