News for hackers. Stuff that matters.
Re:Embed, encode, attend, predict: the new deep learning for... by imh on Friday November 11, 2016 @10:43PM (#12936489)
One other cool part of attention is that you can attend to m-dimensional parts of a n-by-m matrix just as well as a k-by-m matrix. Objects (sentences) of varying size can be treated the same in a really nicely principled way.

There are no comments on this comment yet.