Juan Duque
I’m a third year PhD student in Reinforcement Learning at Université de Montréal & Mila, advised by Aaron Courville. My research unites multi agent RL, game theory, and scalable algorithms, which I apply to social dilemmas as well as climate policy simulators, large language model negotiation, and games. I enjoy turning rigorous theory into efficient algorithms that help AI systems reason and coordinate for real world impact.
Before Mila, I did my master's at the Princeton NLP group with Karthik Narasimhan and my undergraduate at Georgia Tech where I worked with professors Charles Isbell and Michael Loss in Reinforcement Learning.