The Gradient: Perspectives on AI cover image

Joel Lehman: Open-Endedness and Evolution through Large Models

The Gradient: Perspectives on AI

00:00

Is Scaling a Good Idea?

Yes I think that's a really good way of putting it together and one thing I have to wonder since you're speaking to this is really where we're introducing the facet of large models and their influence on these processes. You did report some results from models of different sizes that you experimented with I think you went from like 0.1 million up to 680 million parameters in your pipeline. What what do you found about the size of these models and how that impacted the conditional invention you saw at the end of your pipeline?Yeah it's a good question I'll have to remind myself actually of those those results. I think ultimately on the the RL stage we may not have done much in the way

Play episode from 01:14:07
Transcript

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