Given a probabilistic model, Bayesian inference is straightforward to implement. Building a satisfactory model in a given application, however, is a far more open-ended challenge. In order to ensure robust analyses we need a principled workflow that guides the development of a probabilistic model that is consistent with both our domain expertise and any observed data while also being amenable to accurate computation. In this talk I will discuss recent work that advances a principled workflow for building and evaluating probabilistic models in Bayesian inference.

INSTRUCTIONS

You are not required to bring your laptop to this lecture. Following the lecture, you will get access to a Jupyter notebook that reiterates the material and then guides you through an interactive exercise. To use the notebook you will need to have PyStan (at least version 2.17) installed, https://pystan.readthedocs.io/en/latest/.

Anyone who might want to brush up on Bayesian Inference before the lecture can review the following:

1. https://betanalpha.github.io/assets/case_studies/probability_theory.html
2. https://betanalpha.github.io/assets/case_studies/conditional_probability_theory.html
3. https://arxiv.org/abs/1803.08393