During my DPhil, I have worked on two projects with my supervisors Professor François Caron and Professor Judith Rousseau regarding Bayesian nonparametric models for networks. In particular, we develop models for core-periphery and sparse dynamic networks. We also introduce an approximate inference method, based on a truncation of the infinite-dimensional model.

Image credit: mattbuck (category), CC BY-SA 2.0, via Wikimedia Commons.

I am also working with Professor Weisi Guo, Professor Chenlei Leng, Professor Cassidy Sugimoto and Vincent Larivière on a project to measure the impact of geographic diversity in coauthorships on the citation of their papers across different academic domains. We use the global air transport network in order to construct our diversity measures.

Image credit: The Opte Project, CC BY 2.5, via Wikimedia Commons.

I am currently working with Professor Judith Rousseau and Professor Trevor Campbell on a project concerning Bayesian coresets, in which a small weighted sample is used to approximate a posterior distribution whilst obtaining computational speedups. We propose a novel Bayesian coreset construction algorithm, that improves both the run-time of current methods, and provides more sound theoretical guarantees.

Image credit: Anna Tarazevich from Pexels.

During my MSc, I worked on robust estimation and approximate Bayesian methods in causal inference problems with Professor Emma McCoy and Professor Daniel Graham. Our focus was on doubly and multiply robust methods, with applicatons to road traffic collisions.

Image credit: xkcd.