Mitchell O’Sullivan

PhD, School of Mathematical Sciences

After a four year stint as an analyst modelling payments data to detect and prevent financial crime, Mitchell returned to the Queensland University of Technology in mid-2019 to complete the honours component of his Bachelor of Mathematics.

His project focused on novel sequential Monte Carlo methods for approximate Bayesian computation (ABC). With support from the QUT Centre for Data Science I am now continuing my research efforts in this area as a PhD student focusing on dimensionality reduction for ABC. To progress the field in this area Mitchell aims to develop efficient sequential algorithms that can automatically generate and/or choose summary statistics for complex implicit models.

His research interests include Bayesian statistics, likelihood-free inference, machine learning, and computer science. Mitchell is very enthusiastic about high performance computing and the latest computer hardware.