I’m Phil, a 4th year DPhil/PhD Student at the University of Oxford. My main area of academic interest is improving sample efficiency in reinforcement learning. I’ve also dabbled in generative modelling and algorithmic fairness. I’m supervised by Prof. Steve Roberts in the Machine Learning Research Group. I’ve completed internships at Waymo, investigating interpretibility and uncertainty calibration in deep-learning, and at Microsoft Research, working on continual learning through the lens of network compression. I’ve also been fortunate enough to work with Prof. Sergey Levine and Dr. Ilya Kostrikov as a visiting student at RAIL Lab at UC Berkeley.

Previously I was an AI Resident at Microsoft Research Cambridge where I worked on Deep RL for video games and automatic knowledge base creation. I have Masters Degrees in Machine Learning and Engineering Science from the Universities of Cambridge and Oxford respectively. In a previous life, I was a Management Consultant at Accenture.

In my spare time I like to play the bass guitar and read. I can also speak Mandarin to a relatively fluent level, which has led me to work for extended periods of time in China and Hong Kong.

If you want to email me, please do so at ball AT robots DOT ox DOT ac DOT uk.