TalkRL: The Reinforcement Learning Podcast cover image

Danijar Hafner 2

TalkRL: The Reinforcement Learning Podcast

00:00

Dream Every 4: A Vision for AI

The biggest issue we're facing in our realm now is to do temporal abstraction really well. Another issue is that we want to leverage pre-training data to squeeze out more sample efficiency and learn tasks much faster. Open AI is collecting one, but I doubt they'll share it. And so you have to be data efficient because there just isn't that much data to supervise from. The second part is that we also want our algorithms to adapt quickly. We want these algorithms to discover new things for us.

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