An Effective Approach to Unsupervised Machine Translationの紹介

An Effective Approach to Unsupervised Machine Translationの紹介です。
教師なし翻訳に関するお話です。

Transcript

1. An Effective Approach to Unsupervised Machine Translation
     
3. ?/= 8E 45": 3'209 40G0AIoT< :+;F<%$6 B@-(,F.    !)#7
   *&>12FM2  D1 CD!)#7ML
4. Unsupervised Machine Translation • 87=@16
   Statistical Machine Translation (SMT) Neural Machine Translation (NMT))(9
   5/&%$  ◦ .@.0:2?>! • -B"*< .@;3,A=@4+ ◦ Word translation without parallel data.[Alexis 2017], ◦ Learning bilingual word embeddings with (almost) no bilingual data [Artetxe 2017] • !#'5/ 87=@ NMT>!
   4+ ◦ UNSUPERVISED MACHINE TRANSLATION USING MONOLINGUAL CORPORA ONLY [Lample2018] ◦ Unsupervised statistical machine translation [Artetxe 2018]
5. Supervised Machine Translation   NMT Back-translation  
   !  #"BLEU  http://deeplearning.hatenablog.com/entry/back_translation#f-726c04a7
6. !! • D8?8B;=/@[Alexis 2017] ◦ /@*;="%$#1: ◦ ;=B
   /@)3& A404 6
   - 5.+=A'9C9  7> ◦ +=A( , +=2<EF
7. SMT https://www.nhk.or.jp/strl/publica/rd/rd168/pdf/P14-25.pdf
8. ' 1.  % $ 2. &! 3. SMT$
     " 4. "  refinement 5. NMT(#
9. &9 3+ • bi-gram embedding+A8: #6>$<[Artetxe 2018] •  :  100=0
   / softmax  &952"* (e,f8: 4  :, τ1( ?.',%!7 ) ;-  …@@
10. 2<0K,A • 3N*6 5/2<0KPO   • ex. “Sunday Telegraph” → “The Times of London” • =H. %'#& $"&MQ4 R(8-C WaveNet:1D+@9> 
    IF !) 2<G@7JB; LS 7JE?/ T
11. Unsupervised SMT • Back-translation.CE/;> ◦ DF%"&*8L
    @3 DFB<+4DF%"&.C • 9H7Cycle GAN !#K65= ◦ -:02I   
    ?HA  M 1   : DF'! : ,G(#'$)'! : DF7J'!
12. +% • '$ SMT+% .0 .0 (), +% • SMT+% .0!/1-*&# 
     ()2"
13. NMT$ • "
    SMT$ %# NMT$ • %   NMT#    : SMT%! : NMT%!
14. 
    
15. 
     WMT2014  seq2seq   
16.   
     
    …