Summer school on advanced Bayesian methods

Langeraert, W. (Participant), Lommelen, E. (Participant), Verschelde, P. (Participant), Van Calster, H. (Participant), Carmen, R. (Participant)

Activity: Participating in or organising an event types › Organisation and participation in conference, workshop, training, seminar, meeting

Description

The Interuniversity Institute for Biostatistics and statistical Bioinformatics organizes for the 5th time the Summer School on Advanced Bayesian Methods.

The course is given by Dr. Charles Margossian (Flatiron Institute, Center for Computational Mathematics in New York)

"In this course, we will discuss the tenants of the Bayesian workflow and how to execute them using state-of-the-art software. The Bayesian workflow is the iterative process through which we build, fit, and criticize models, with the latter step often motivating useful revisions to our model. We will take advantage of Stan, a Bayesian inference software which boasts a flexible language to specify models, and supports scalable inference algorithms, notably an adaptive Hamiltonian Monte Carlo (HMC) sampler — currently one of the most successful Markov chain Monte Carlo (MCMC) methods. This will be a hands-on workshop: students will be expected to code and attempt several exercises. Throughout, ongoing research as well as open questions on the subject of Bayesian modeling will also be highlighted."

Objectives of the course are:

Learn the Stan language
-Develop a (deeper than usual) understanding of Bayesian inference using MCMC
-Learn the fundamentals of the Bayesian workflow
-Apply these principles to hierarchical models and ODE-based models, with an optional session on applications in pharmacometrics using the add-on Torsten
Period11-Sep-2023 → 13-Sep-2023