Episode 22: Archit Sharma, Stanford, on unsupervised and autonomous reinforcement learning
Generally Intelligent
00:00
Scaling Robotics
Are there areas of research that are underinvested or bottlenecks where you want people to work on them more? I would definitely think that some of the research I'm doing with people can look into it a bit more. It's freaking like create systems which require a very little supervision. Maybe we can ever hundreds or thousands of robots straight across the world just like learning.
Transcript
Play full episode
