Jeff Dean: evaluation

Transcript context

…It feels like the algorithms are following the hardware. Basically, what's happened is that at this point, arithmetic is very, very cheap, and moving data around is comparatively much more expensive. So pretty much all of deep learning has taken off roughly because of that. You can build it out of matrix multiplications that are N cubed operations and N squared bytes of data communication basically. Well, I would say that the pivot to hardware oriented around that was an important transition, because before that, we had CPUs and GPUs that were not especially well-suited for deep learning. And then we started to build TPUs at Google that were really just reduced-precision linear algebra machines, and then once you have that then you want to exploit it. It seems like it's all about identifying opportunity costs. Like, okay, this is something like Larry Page, I think, used to always say: "Our second biggest cost is taxes, and our biggest cost is opportunity costs." If he didn't say that, then I've been misquoting him for years. But basically it’s like, what is the opportunity that you have that you're missing out on? In this case, I guess it was that you've got all of this chip area, and you're putting a very small number of arithmetic units on it. Fill the thing up with arithmetic units! You could have orders of magnitude more arithmetic getting done. Now, what else has to change? Okay, the algorithms and the data flow and everything else.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.