Episode 22: Archit Sharma, Stanford, on unsupervised and autonomous reinforcement learning
Generally Intelligent
The Variational Empowerment Paper
Unsupervised RL is like filling out supervised. Like you just need to learn which things are not reversible in your environment. I think that's a really good way of thinking about it. And one of the really cool things about this is that it's fully unsupervised. They actually have some work which does that. It was very clever idea because it was actually sent unsupervised as well.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
