Hippolyt Ritter

About me

I am a machine learning researcher. I most recently did a post-doc at Meta in the Probability org with Theofanis Karaletsos. Prior to that I completed my PhD at University College London advised by David Barber. My thesis was titled “Scalable approximate inference methods for Bayesian deep learning”. During my PhD, I did internships at Microsoft Research Cambridge with Yingzhen Li and Cheng Zhang, and at Uber AI Labs with Theofanis Karaletsos.

Even before that, I completed an MSc at University College London in Computational Statistics and Machine Learning. I did my BSc in Bioinformats at the LMU and TU in Munich.

Research interests

I am broadly interested in probabilistic machine learning and more specifically Bayesian deep learning. My work has revolved around developing structured posterior approximations that are efficient enough to apply to large neural networks. I am particularly interested in settings that go beyond uncertainty estimation, such as continual learning or data distillation. I strongly believe in making research work more broadly accessible and have led the development of TyXe, an open source library for Bayesian neural networks based on Pyro.

Selected publications

Thesis

Scalable approximate inference methods for Bayesian deep learning
PhD thesis, University College London, 2023.

Powered by Jekyll and Minimal Light theme.