An approach used for controlling the lower body of the robot with a reinforcement learning policy.
Stanford Online
