PhD Student
leedsharkey[at]gmail[dot]com
Google scholar
Personal website
@leedsharkey
CV (pdf)
Lee is interested in understanding what is going on inside cognitive systems in order to ensure that intelligent machines are safe. Concretely, he focuses on gaining a mechanistic interpretation of what neural networks have learned. He draws on tools and theory from cognitive science, machine learning (especially deep reinforcement learning), and computational neuroscience. He did a Bachelor’s degree in preclinical medical studies at the University of Cambridge, where he majored in neuroscience. He dropped out of medical school to work for several years in international public health, which exposed him to the deeply tangled web of humanitarian problems that the world faces. Discontent with working on only one of those problems, and after a philosophical shift toward long-termism, he switched career to study artificial intelligence. This switch motivated his MSc degrees in Data Analytics (University of Glasgow) and Neural Systems and Computation (University of Zurich/ETH Zurich) and his current research PhD here in the Human and Machine Cognition lab.