Hengyuan Hu

profile photo

I am a PhD student at Stanford University advised by Dorsa Sadigh. I am interested in robotics, human-AI coordination and reinforcement learning.

Previously, I spent 4 years as a Research Engineer at FAIR where I worked closely with Jakob Foerster and Noam Brown on reinforcement learning, search and coordination in multi-agent environments. I was a core member of the team that created Cicero, the first AI that reaches human level in Diplomacy. I also trained the most human-like agent in Hanabi without using any human data.

Email  /  Github  /  Google Scholar

Robotics

Human-AI Coordination

Learning Diverse Policies

Others