Recurrent Neural Network Regularization
lastmod 2018-08-17created 2016-10-282 edits
This paper proposes a way to make Dropout work well by applying Dropout to LSTM in a form that makes good use of the characteristics of RNN.
Local Download
- RECURRENT NEURAL NETWORK REGULARIZATION (arXiv - 1409.2329)
- https://arxiv.org/pdf/1409.2329v5.pdf
-
1409.2329v5.pdf
See also
Favorite site
References
-
Sanghyukchun_-Recurrent_Neural_Network_Regularization-_README.pdf ↩