The Future of AI

The idea of being able to interact with the same theme, basically, of limiting the amount of supervision that is needed to learn the interaction. One of Deepak's most recent results where the system can learn how to have a person on for the robot with just one camera. No fancy gloves, motion captures, 3D cameras, you know, lots of these modes and all this,. But the interesting thing is that that is learned entirely from completely unannotated, completely raw videos from YouTube.