Research and AI Institute (June 2024 -Jan 2025)

I trained Spot, a quadruped robot, to juggle a ping-pong ball through Reinforcement Learning. Subsequently, I developed Diffuse-cLoC, a guided diffusion model for physics-based character control.