TalkRL: The Reinforcement Learning Podcast cover image

Jeff Clune

TalkRL: The Reinforcement Learning Podcast

00:00

The Goal of Open Endowed Learning Systems

In general, I do agree with you evolution doesn't really have a goal. People didn't set out to produce ants or Cheetos or humans. And the same thing is going to be true of our algorithms. We're going to have to create systems that are not seeking a goal. That will ultimately unlock tremendous progress. But as you said, how do I know when it's worth running for a lot longer? This is very difficult to measure, but often we can just kind of look in there and see if really interesting things are happening then we know that it's worth continuing.

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