Episode 19: Minqi Jiang, UCL, on environment and curriculum design for general RL agents
Generally Intelligent
00:00
Is There Something Big Missing?
Orto: I think one thing that sort of bothers me about existing things is like they always just, it's alla pe of next action. He says we don't really have a weight even express to day. One promising in road to that is looking at language models. Orto: If you can get transformer models working very well, that trained on off lind trajectory data, it seems like a natural extension of that idea to basically d liht forward prediction.
Transcript
Play full episode
