Episode 18: Oleh Rybkin, UPenn, on exploration and planning with world models
Generally Intelligent
Theorems of Theoeo Fasti
Most of the time we learn to do things, just badly. And a lot of our current agents are kind of free in but morallys thinking. But what would be much better if you had a fully un supervised agent? The paper i'm currently most excited about is called 'Plan to explore' It's a scaled i version of plan to explore that cand of fix us sum of the isias.
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.
