I am a Senior Research Scientist at DeepMind in London, England. I completed my PhD research in 2016 under Yoshua Bengio at Université de Montréal as part of (what shortly thereafter became) Mila. Prior to that, I completed Bachelor's and Master's degrees at the Department of Computer Science at the University of Toronto, the latter supervised by Quaid Morris.
My work currently focuses on reinforcement learning in sparse reward settings, more specifically unsupervised/self-supervised objectives for control. I am also interested in core deep learning methods, in particular unsupervised representation learning and likelihood-free approaches to generative models.
You can find out more about my past research from Google Scholar, and find some code of varying vintages that I've written on my (mostly inactive) GitHub page. You can also find me on Twitter and LinkedIn.