About

Since fall 2022, I am a postdoc in Professor Tamara Broderick’s group at the MIT Laboratory For Information and Decision Systems. Previously, I was a PhD student in the Machine Learning Group at the University of Cambridge, supervised by Professor Carl Edward Rasmussen.

Research Interests

I develop methods for validation and uncertainty quanitifcation for spatiotemporal data. When data has spatiotemporal structure, commonly used approaches for validation make unreasonable assumptions, such as that the data is independent and identically distributed. These unreasonable assumptions often lead to validation approaches that overstate confidence in conclusions reached using machine learning approaches. I develop methods that replace i.i.d. assumptions with more physically interpretable assumptions. I have also worked extensively on approximate inference in Gaussian processes and other Bayesian models.