Colin Lu

Hi, I’m Colin! I’m starting my PhD at MILA/University of Montreal with Aaron Courville in Fall 2026. My main focus is multi-agent RL, especially with a game-theoretic flavor. I’m also interested in single-agent RL, interpretability, and safety/alignment, among other things.

Previously, in my gap years I worked with Jakob Foerster and Shimon Whiteson at Oxford on MARL/opponent shaping, and with Ben Eysenbach at Princeton on goal-conditioned RL, exploration, and hierarchical RL. In 2024 I completed my undergrad at UIUC, double majoring in CS and statistics, and minoring in math.