Episode 28: Sergey Levine, UC Berkeley, on the bottlenecks to generalization in reinforcement learning, why simulation is doomed to succeed, and how to pick good research problems
Generally Intelligent
00:00
General and Reusable Robotic Control Models
Researchers have been working on how to build general and reusable robotic control models. One of their results that's kind of closest to this is a paper by Drew Shaw called general navigation models. The model could actually figure out how to fly the drone immediately, he says.
Transcript
Play full episode
