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
Is It Sufficient to Just Build Larger Language Models?
There are a couple of things that I could say on this topic. Is it sufficient to simply build larger language models? You need the ability to learn patterns in data, any of the ability to plan. Planning really refers to some kind of computational process that determines the course of action. It doesn't necessarily need to be literally like you think of individual steps in a plan.
Transcript
Play full episode
