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
Computer Graphics Research
In 2009, we already had pretty good technology for simulating physics. And really the big challenge was getting plausible behavior out of virtual humans. Once I realized how big that problem really was, it became natural to think of it more as just developing artificial intelligence systems.
Transcript
Play full episode
