
Episode 25: Nicklas Hansen, UCSD, on long-horizon planning and why algorithms don't drive research progress
Generally Intelligent
00:00
Is There a Lot of Research in ML Models?
I think there is a lot of still intuition missing in RL. Like when are algorithms when you train it? When is it supposed to be successful? And when when it fail? I think more systematic categorization of this would save collectively as a community a lot of time. Dan: It almost argues for like open source robotic platform that we sort of standardized on, so everyone can get the same results.
Transcript
Play full episode