User Tools

Site Tools


nlp:lstm

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
nlp:lstm [2021/02/18 13:38] jmflanignlp:lstm [2023/06/15 07:36] (current) – external edit 127.0.0.1
Line 8: Line 8:
  
 <blockquote> <blockquote>
-Conneau et al. (2017) explore multiple different sentence embedding architectures+[[https://arxiv.org/pdf/1705.02364.pdf|Conneau et al. (2017)]] explore multiple different sentence embedding architectures
 ranging from LSTM, BiLSTM and intra-attention to convolution neural networks ranging from LSTM, BiLSTM and intra-attention to convolution neural networks
 and the performance of these architectures on NLI tasks. They show that, out of and the performance of these architectures on NLI tasks. They show that, out of
Line 15: Line 15:
 performance on various transfer learning tasks. performance on various transfer learning tasks.
 </blockquote> </blockquote>
 +
 +With tweaks, they can outperform Transformer models.  See [[https://arxiv.org/pdf/1804.09849.pdf|Chen et al 2018 - The Best of Both Worlds: Combining Recent Advances in Neural Machine Translation]].
  
 ===== Resources ===== ===== Resources =====
   * [[https://people.idsia.ch/~juergen/lstm/|Jeurgen's LSTM Tutorial]]   * [[https://people.idsia.ch/~juergen/lstm/|Jeurgen's LSTM Tutorial]]
nlp/lstm.1613655527.txt.gz · Last modified: 2023/06/15 07:36 (external edit)

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki