Joe Bacchus George
Probabilistic Systems, Information and Inference Group
University of Cambridge
jbg39@cam.ac.uk
I am a PhD student at the University of Cambridge working under Dr. George Cantwell and Prof. Ramji Venkataramanan. I am a member of Trinity College.
I am interested in complex systems, and believe that the best way to do this is by using probabilistic models and using statistical inference to answer questions about them.
Complexity usually relates to systems with lots of interactions, so we tend to model them as graphs. It’s usually impossible to know all the intricate details about these systems, so adopting a Bayesian framework allows us to remain transparent about our uncertainties. Even when we propose a simple model, the questions we like to ask about them tend to remain intractable, both analytically and computationally. For this reason, I am most interested in methods for performing approximate inference in these graph-based settings, like message-passing and mean-field methods.
I previously completed a master’s in mathematical and theoretical physics at the University of Oxford, and a bachelor’s in theoretical physics at University College London (UCL).
