
Episode 13: Jonathan Frankle, MIT, on the lottery ticket hypothesis and the science of deep learning
Generally Intelligent
00:00
Object Oriented Programming
I encourage folks keep a balance. Ti, let's eas let's make some progress. Let's keep doing good science. We'll see were obviously at the beginning. It's only meant five years. No, chel, maybe ben, a little bit longer. A people have been doing machine long i think one of the points that made in one of your papers peoplve been pruning networks and looking at sparsity for a really, really long time. None of this stuff is new. There's some very old methods that have been reinvented. I agree with the li keepit middle guy's not following the guysbout exploding. And i think we can move ourselves towards
Transcript
Play full episode