Generally Intelligent cover image

Episode 28: Sergey Levine, UC Berkeley, on the bottlenecks to generalization in reinforcement learning, why simulation is doomed to succeed, and how to pick good research problems

Generally Intelligent

00:00

Reinforcement Learning - The Basic Problem With Offline RL

In the past, people thought that offline RL really wasn't that different from kind of traditional value-based methods like Q-learning. But now is a very widely accepted notion that this distributional shift is a very fundamental challenge in offline reinforcement learning. And it really deeply connects to counterfactual inference. Reinforcement learning is really not kind of factual. It's about saying, well, I saw you do this, and that was the outcome. What if you did something different? Would the outcome be better or worse? That's the basic question that Reinforcer Learning asks.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app