Paul Festor

PhD Student

Paul is a PhD student from the AI4Health CDT. His research focuses mainly on machine learning safety, with a focus on reinforcement learning, to deploy clinical decision support systems at the bedside. He got his engineer degree from the “Ecole des Mines de Nancy” (France) and his MSc degree in Computing with a specialisation in Ai and Machine Learning from Imperial College London.

Research Brief

Paul is interested in bringing machine learning to the real world in a safe way. He focuses mainly on off-policy reinforcement learning, probabilistic modelling and uncertainty estimation applied to clinical decision support systems.

Lab Publications

Assuring the safety of AI-based clinical decision support systems: a case study of the AI Clinician for sepsis treatment


Paul Festor, Yan Jia, Anthony C Gordon, A Aldo Faisal, Ibrahim Habli, Matthieu Komorowski

BMJ health & care informatics

MIDAS: Deep learning human action intention prediction from natural eye movement patterns


Paul Festor, Ali Shafti, Alex Harston, Michey Li, Pavel Orlov, A Aldo Faisal

arXiv preprint arXiv:2201.09135

Levels of Autonomy and Safety Assurance for AI-Based Clinical Decision Systems


Paul Festor, Ibrahim Habli, Yan Jia, Anthony Gordon, A Aldo Faisal, Matthieu Komorowski

International Conference on Computer Safety, Reliability, and Security

Deep learning human action intention classification from natural eye movement patterns


Ali Shafti, Paul Festor, Pavel Orlov, Mickey Li, A Aldo Faisal

Journal of Vision

Enabling risk-aware Reinforcement Learning for medical interventions through uncertainty decomposition


Paul Festor, Giulia Luise, Matthieu Komorowski, A Aldo Faisal

IMLH - ICML 2021