Decomposing the Generalization Gap in Imitation Learning for Visual Robotic Manipulation

Open in new window