Tom Bush

Hi, I’m Tom! I am currently researching learning planning in model-free deep RL agents as a research assistant in Cambridge’s Krueger AI Safety Lab. I am supervised by Usman Anwar and Stephen Chung. I have just finished my MPhil in Machine Learning and Machine Intelligence at the University of Cambridge, before which I studied Philosophy, Politics and Economics at the London School of Economics.

I am interested in ensuring advanced AI lives up its potential to do good. I am especially interested in topics relating to “AI safety” such as interpretability and robustness. One related research direction I am particularly focused on at the moment is applying interpretability tools to decode learned analogues to human reasoning processes within RL agents. On the other hand, however, I am also fascinated by applying AI to solve important problems. I will soon be looking for a more permanent position, so definitely reach out if you are interested in either of these fields!