Special PIMS-UBC Math Colloquium/Probability Seminar: Jonathan Mattingly

I will begin with a brief description of how computational sampling methods have been critical in developing methods to detect political maps which are gerrymandered. I will discuss some of the court cases and political implications in US elections. I then will dive into some of the algorithms we use and some of the mathematical difficulties we have encountered. I will talk about random walks on spanning forests, characterizing high-dimensional phase spaces, parallel tempering, and Markov Chain Monte Carlo.