I am an artificial intelligence (AI) PhD student at UC Berkeley, supervised by Stuart Russell. My goal is to develop techniques necessary for advanced automated systems to verifiably act according to human preferences, even in situations unanticipated by their designer. I am particularly interested in improving methods for value learning and preference elicitation, such as inverse reinforcement learning.
I work closely with the Center for Human-Compatible AI, a cross-disciplinary research centre. Prior to joining Berkeley, I had the pleasure of working with Zoubin Ghahramani and Christiain Steinruecken during my Master’s degree in the Machine Learning Group at the University of Cambridge. Please see my CV for a more comprehensive list of my prior experience.