Skip to content

Latest commit

 

History

History
10 lines (7 loc) · 511 Bytes

rnn-regularization.md

File metadata and controls

10 lines (7 loc) · 511 Bytes

TLDR; The authors show that applying dropout to only the non-recurrent connections (between layers of the same timestep) in an LSTM works well, improving the scores on various sequence tasks.

Data Sets and model performance

  • PTB Language Modeling Perplexity: 78.4
  • Google Icelandic Speech Dataset WER Accuracy: 70.5
  • WMT'14 English to French Machine Translation BLEU: 29.03
  • MS COCO Image Caption Generation BLEU: 24.3