Re:New kind of recurrent neural network using attention
by jostmey on Wednesday March 08, 2017 @01:09PM
(#13822179)
I have not tested GRU cells on these problems. The GRU cells are not too different from LSTM cells--both models make heavy use of gating terms. As such, I don't expect a GRU model to do much better than a LSTM. But I cannot say for certain, having not tried it.
[ Parent
| on HN ]
There are no comments on this comment yet.