News for hackers. Stuff that matters.
Re:No training required: Exploring random encoders for sente... by anon1253 on Wednesday January 30, 2019 @03:24PM (#19039048)
Love it. Especially the echo state network trick. I wonder how much of BERT/ELMO performance is simply due to them having a such a high dimensionality. Not that there is anything wrong with that, just makes a tad less practical for some applications.

There are no comments on this comment yet.