Statistical machine learning
University of Oxford | Oxford-Man Institute
About me
g.duran@me.com
I am a postdoctoral researcher at the Oxford-Man Institute, University of Oxford.
My work develops probabilistic methods that allow intelligent agents learn and adapt continually in uncertain, dynamic environments — especially when data is scarce or noisy. I use tools from filtering and Bayesian inference to tackle challenges in online continual learning, neural contextual bandits, reinforcement learning, and time-series forecasting.
In short: I study how to build AI agents that never stop learning.
Some talks
- September 2025 | MatheEXLab Webinar @ NUS | OnlineFrom spaceships to neural networks: Bayesian filtering as the language of online continual learning
- June 2025 | Bayescomp Satellite Workshop on misspecified models | SingaporeA unifying framework for generalised Bayesian online learning in non-stationary environments
- June 2025 | UCL ELLIS CSML Seminar | EnglandA unifying framework for generalised Bayesian online learning in non-stationary environments