News for hackers. Stuff that matters.
Re:How a Toronto professor’s research revolutionized artific... by whitten on Friday April 17, 2015 @04:33PM (#9397049)
Is this news because "neural networks" now has a new buzzword "deep learning" and Google is paying attention to it?
Re:How a Toronto professor’s research revolutionized artific... by Animats on Friday April 17, 2015 @04:43PM (#9397101)
It's news because that's the cold place neural networks went to wait out the "AI winter".

It's amazing that neural networks actually work now. They've been around since the 1960s, and for four decades, they sucked. They're only a little smarter; it's mostly sheer compute power applied to algorithms that fit on one screen of Matlab.

Re:How a Toronto professor’s research revolutionized artific... by murbard2 on Friday April 17, 2015 @04:56PM (#9397161)
Not so much the computing power as a few tricks:

- proper initialization of weights (should have been obvious, but no one really saw it)

- rectified linear units and dropout

Re:How a Toronto professor’s research revolutionized artific... by UmDieWelt on Friday April 17, 2015 @05:27PM (#9397285)
Also, a lot more data to train on.
Re:How a Toronto professor’s research revolutionized artific... by frandroid on Friday April 17, 2015 @05:37PM (#9397338)
$400M+ is a lot of attention.