Episode 13: Jonathan Frankle, MIT, on the lottery ticket hypothesis and the science of deep learning
Generally Intelligent
00:00
Deep Learning
A lot of the research we do in deep learning is much more o the engineering style. It's a hard thing to do, though, any kind of research where you're just trying to make something happen. You may be working on a problem impossible, and you have no idea. And maybe some fundamental new insight on the world will just unlock that for free. I've probably spent as much energy as anybody on it. And i wish it had been successful.
Transcript
Play full episode
