
In Machines We Trust: A conversation with Geoffrey Hinton (Live)
MIT Technology Review Narrated
00:00
Back Propagation in Large Language Models
The technique basically performed really well on ImageNet. And we had Joe Elpino from Meta yesterday showing how far image detection had come. And it's also the technique that underpins large language models. So I want to talk now about this technique, which you initially were thinking of as almost like a poor approximation of what biological brains might be. Yes. There's no one in the audience here that's smiling and thinking that was a silly explanation.
Transcript
Play full episode