Yann LeCun explains why Facebook would crumble without AI
The Robot Brains Podcast
How to Kill Nearall Networks With a Book?
The main limitation was the fact that the perception basically has a single layer of tranable weights, right? So you can think of the perception on vasicaly as a single noron with, you know, a bunch of adjustable inputs. And what the recent brand of noral nets allowed to do was to train noral nets with multiple layers of adjustable weights. Is called deep running for the simple fact that those noral nets are composed of multiple layers.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
