Ilya Sutskever explains the origins of deep learning
The Robot Brains Podcast
00:00
Neural Networks
Ilia: I had very little doubt that it would work at this point. Alex was straining the newral net, and the results were getting better week after week. But i felt like the big risk, for my perspective, was, do we have the ability to utilize the gpus well enough? No, big enough? Thereis no such thing. It has to be a neural network that is large enough to be interesting. If you're just going to have something which is going to be weigh larger than anything before, then it should do much better than anything anyone's ever seen. Of course, we ar far beyond that. Our computers are faster, and ou networks are
Transcript
Play full episode
