Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Speaker Deck
PRO
Sign in
Sign up for free
Soft_Contextual_Data_Augmentation_for_Neural_Machine_Translation_.pdf
MARUYAMA
August 25, 2019
0
61
Soft_Contextual_Data_Augmentation_for_Neural_Machine_Translation_.pdf
MARUYAMA
August 25, 2019
Tweet
Share
More Decks by MARUYAMA
See All by MARUYAMA
vampire.pdf
tmaru0204
0
69
Misspelling_Oblivious_Word_Embedding.pdf
tmaru0204
0
52
Simple_Unsupervised_Summarization_by_Contextual_Matching.pdf
tmaru0204
0
62
Controlling_Text_Complexity_in_Neural_Machine_Translation.pdf
tmaru0204
0
61
20191028_literature-review.pdf
tmaru0204
0
63
Hint-Based_Training_for_Non-Autoregressive_Machine_Translation.pdf
tmaru0204
0
46
An_Embarrassingly_Simple_Approach_for_Transfer_Learning_from_Pretrained_Language_Models_.pdf
tmaru0204
0
56
Addressing_Trobulesome_Words_in_Neural_Machine_Translation.pdf
tmaru0204
0
48
Simple_Unsupervised_Keyphrase_Extraction_using_Sentence_Embeddings.pdf
tmaru0204
0
85
Featured
See All Featured
Build The Right Thing And Hit Your Dates
maggiecrowley
19
1.2k
JazzCon 2018 Closing Keynote - Leadership for the Reluctant Leader
reverentgeek
172
8.3k
Code Review Best Practice
trishagee
41
6.8k
The Invisible Customer
myddelton
110
11k
Building a Scalable Design System with Sketch
lauravandoore
447
30k
Fantastic passwords and where to find them - at NoRuKo
philnash
26
1.5k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
350
21k
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
15
920
Teambox: Starting and Learning
jrom
121
7.6k
Clear Off the Table
cherdarchuk
79
280k
5 minutes of I Can Smell Your CMS
philhawksworth
196
18k
Creatively Recalculating Your Daily Design Routine
revolveconf
207
10k
Transcript
4PGU$POUFYUVBM%BUB"VHNFOUBUJPO GPS/FVSBM.BDIJOF5SBOTMBUJPO 'FJ(BP +JOIVB;IV -JKVO8V :JOHDF9JB 5BP2JO 9VFRJ$IFOH 8FOHBOH;IPV 5JF:BO-JV
"$- QBHFTr -JUFSBUVSFSFWJFX /BHBPLB6OJWFSTJUZPG5FDIOPMPHZ5BLVNJ.BSVZBNB
"CTUSBDU ⾣จ຺Λߟྀͨ͠ιϑτͳσʔλ֦ு ⾣χϡʔϥϧػց༁ʹ͓͚Δσʔλ֦ுख๏ΛఏҊ ⾣σʔλ͕গͳ͍߹ɾେྔʹ͋Δ߹ɺͲͪΒʹ͓͍ͯ طଘͷσʔλ֦ுख๏Λ্ճΔੑೳΛୡ
.FUIPE ⾣4PGUDPOUFYUVBMEBUBBVHNFOUBUJPO ɾFNCFEEJOHPGTPGUXPSEТ EFNCFEEJOHNBUSJY
&YQFSJNFOU ⾣#BTFMJOFT #BTFσʔλ֦ுख๏ෆ༻ 4XBQ૭෯LͷதͰɺϥϯμϜʹ୯ޠͷҐஔΛೖΕସ͑ %SPQPVUϥϯμϜʹ୯ޠΛܽམ
&YQFSJNFOU ⾣#BTFMJOFT #MBOLϥϯμϜʹQMBDFIPMEFSUPLFOl@zʹஔ͖͑ 4NPPUIVOJHSBNͷසʹج͍ͮͯ୯ޠΛαϯϓϦϯά͠ɺ -.TBNQMF ݴޠϞσϧͷग़ྗʹج͍ͮͯ୯ޠΛαϯϓϦϯά͠ɺ ϥϯμϜʹஔ͖͑ ϥϯμϜʹஔ͖͑
&YQFSJNFOU ⾣%BUBTFUT ɾ*845\(FSNBO 4QBOJTI )FCSFX^ˠ&OHMJTI ɾ8.5&OHMJTIˠ(FSNBO ⾣.PEFM ɾ-. /.5USBOTGPSNFSBSDIJUFDUVSF ɾSFQMBDJOHQSPCBCJMJUZЍ\
^
&YQFSJNFOU
&YQFSJNFOU ఏҊख๏ЍͰɺ #-&6͕࠷େ طଘख๏ЍͰɺ #-&6Լ
$PODMVTJPOT ⾣จ຺Λߟྀͨ͠ιϑτͳσʔλ֦ு ⾣χϡʔϥϧػց༁ʹ͓͚Δσʔλ֦ுख๏ΛఏҊ ⾣σʔλ͕গͳ͍߹ɾେྔʹ͋Δ߹ɺͲͪΒʹ͓͍ͯ طଘͷσʔλ֦ுख๏Λ্ճΔੑೳΛୡ