Generally Intelligent cover image

Episode 21: Chelsea Finn, Stanford, on the biggest bottlenecks in robotics and reinforcement learning

Generally Intelligent

CHAPTER

Resonance Learning - Sample Efficiency

We used to be really excited about object-centric models also. But then thought about it a lot more and it's like, well, what is an object? Is the brick an object? Well, the wall. What about this plan? Yeah. Objects are fractal in a way in the real world. And we form representations of objects in some other way. That is not how theseobject-centric models are learning. One also should be based off of the tasks that you're doing. If you want to ride a bike, you model the bike as a bike. Whereas if you want to fix a bike or fix a flat tire, you're going to model individual parts of the

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.
App store bannerPlay store banner