Natasha Jaques 2
TalkRL: The Reinforcement Learning Podcast
00:00
Social Learning Versus Imitative Learning
Sci-fi learning is about how to learn from other agents in your environment. It's a little, it's kind of uses the math of successor features so maybe a little hard to describe on a podcast. And those social learning agents end up generalizing a lot better than agents that are trained with imitation learning or with RL,. So I think that whole direction is actually quite interesting if you're just trying to discover everything.
Transcript
Play full episode
