News for hackers. Stuff that matters.

A comment on Deep learning has a size problem. Parent comment

Re:Deep learning has a size problem by kyle_grove on Saturday November 09, 2019 @07:25PM (#21495531)
In fact, one of the chief advantages of the BERT/Transformer architecture over ELMO/LSTM is the ability to parallelize.
[ Parent | on HN ]

There are no comments on this comment yet.