Bristol Bayes for the Brain Workshop

11:00-12:00 13 June 2022 - Hong Ge

Automating Bayesian workflows for basic science

Abstract: Bayesian modelling and inference provide a rigorous framework for analysing and interpreting data, informing decision-making and creating intelligent agents that can learn from experience. However, it is often analytically challenging and error-prune to design and deploy Bayesian algorithms to real-world problems. This talk will walk through a few examples and illustrate how a probabilistic programming software Turing can help automate some of the complex steps in designing and applying Bayesian models.

Brief bio: Hong is a Senior Research Fellow at the Department of Engineering, University of Cambridge. He obtained his Ph.D. (advised by Zoubin Ghahramani) and did postdoctoral work at the University of Cambridge. After his PhD and postdoc, he became a Senior Research Fellow at the same group. His research interests are in theoretical and computational principles of learning and intelligence. He develops novel models as well as efficient algorithms for inference and learning. He created the Turing language for probabilistic programming.