Jamie Reason

Jamie Reason

PhD student, Stats and ML, University of Oxford

I am a PhD student on the StatML CDT at the University of Oxford, supervised by Mark van der Wilk and Chris Holmes. I am interested in how machine learning systems can make good decisions as data is observed over time. My work focuses on amortised Bayesian inference with neural processes and tabular foundation models, making their beliefs reliable enough to act on and extending them to more challenging settings. This includes adapting how much computation they spend, maintaining beliefs that can be updated rather than recomputed, and pooling information across many units observed in parallel.

A non-exhaustive list of topics I am interested in:

Publications

Jamie Reason, Vik Shirvaikar, Mark van der Wilk, Chris Holmes