
Episode 18: Oleh Rybkin, UPenn, on exploration and planning with world models
Generally Intelligent
00:00
I've Been Thinking About Reason
When i went to school, i originally was going to do competational neroscience. But i felt exactly the same way fout like all this isn't really telling me anything about intelligence or learning. Sothat i wonder if we might end up making models that are very powerful and potentially intelligent that we just don't understand at all. And for instance, one real interesting question is, will what if we just take the largest transformer in existence and we try an aton the detiset of all human experience? Well, we then get human level a rigt.
Transcript
Play full episode