Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
文献紹介: 直訳性を利用した機械翻訳知識の自動構築
Search
Yumeto Inaoka
April 26, 2017
Technology
0
140
文献紹介: 直訳性を利用した機械翻訳知識の自動構築
2017/04/26の文献紹介で発表
Yumeto Inaoka
April 26, 2017
Tweet
Share
More Decks by Yumeto Inaoka
See All by Yumeto Inaoka
文献紹介: Quantity doesn’t buy quality syntax with neural language models
yumeto
1
190
文献紹介: Open Domain Web Keyphrase Extraction Beyond Language Modeling
yumeto
0
240
文献紹介: Self-Supervised_Neural_Machine_Translation
yumeto
0
160
文献紹介: Comparing and Developing Tools to Measure the Readability of Domain-Specific Texts
yumeto
0
170
文献紹介: PAWS: Paraphrase Adversaries from Word Scrambling
yumeto
0
160
文献紹介: Beyond BLEU: Training Neural Machine Translation with Semantic Similarity
yumeto
0
280
文献紹介: EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing
yumeto
0
340
文献紹介: Decomposable Neural Paraphrase Generation
yumeto
0
230
文献紹介: Analyzing the Limitations of Cross-lingual Word Embedding Mappings
yumeto
0
230
Other Decks in Technology
See All in Technology
【5分でわかる】セーフィー エンジニア向け会社紹介
safie_recruit
0
30k
Grafana Meetup Japan Vol. 6
kaedemalu
1
200
Skrub: machine-learning with dataframes
gaelvaroquaux
0
110
新規案件の立ち上げ専門チームから見たAI駆動開発の始め方
shuyakinjo
0
640
生成AI時代のデータ基盤
shibuiwilliam
4
2k
Language Update: Java
skrb
2
180
つくって納得、つかって実感! 大規模言語モデルことはじめ
recruitengineers
PRO
32
12k
Bye-Bye Query Spaghetti: Write Queries You'll Actually Understand Using Pipelined SQL Syntax
tobiaslampertlotum
0
120
サンドボックス技術でAI利活用を促進する
koh_naga
0
150
データアナリストからアナリティクスエンジニアになった話
hiyokko_data
0
260
JavaScript 研修
recruitengineers
PRO
6
1.4k
実運用で考える PGO
kworkdev
PRO
0
130
Featured
See All Featured
Building Applications with DynamoDB
mza
96
6.6k
Mobile First: as difficult as doing things right
swwweet
224
9.9k
What's in a price? How to price your products and services
michaelherold
246
12k
Designing for Performance
lara
610
69k
BBQ
matthewcrist
89
9.8k
Thoughts on Productivity
jonyablonski
69
4.8k
Measuring & Analyzing Core Web Vitals
bluesmoon
9
570
Understanding Cognitive Biases in Performance Measurement
bluesmoon
29
1.9k
Reflections from 52 weeks, 52 projects
jeffersonlam
351
21k
Build your cross-platform service in a week with App Engine
jlugia
231
18k
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
252
21k
ReactJS: Keep Simple. Everything can be a component!
pedronauck
667
120k
Transcript
༁ੑΛར༻ͨ͠ػց༁ࣝͷ ࣗಈߏங จݙհ ࣗવݴޠॲཧݚڀࣨɹҴԬເਓ ࠓଜݡ࣏ ۱ాӳҰ দຊ༟࣏ ࣗવݴޠॲཧ
7PM /P QQ
എܠ ˗ ༁ͷଟ༷ੑؚ͕·ΕΔίʔύεʹΑΔػց༁ ˠͳࣝͷ֫ಘʹΑΔޡ༁ͷݪҼ ˗ ػց༁ʹదͨ͠ର༁ʹݶఆ ˠ༁࣭ͷ্ ˗ ੍ݶର༁ͷࢦඪͱͯ͠༁ੑΛར༻
༁ͷଟ༷ੑ ˗ จ຺ঢ়گʹґଘͨ͠༁ ྫ จจ຺ͰղܾͰ͖ͳ͍ᐆດͳUIF ʮࢲͷʯʮͦͷʯͷݶఆදݱ͕༩ ྫ ʮࣸਅΛࡱ͍͚ͬͯͨͩ·͔͢ʯ
ˠl$PVMEZPVUBLFPVSQIPUPHSBQI z ˠl$PVMEZPVQSFTTUIJTTIVUUFSCVUUPO z
༁ͷଟ༷ੑ ˗ ݴ͍͑දݱ ྫ ʮ͜ͷτϥϕϥʔζνΣοΫΛݱۚʹ͍ͯͩ͘͠͞ʯ ˠl*`EMJLFUPDBTIUIFTFUSBWFMMFS`TDIFDLTz ฏঀจ ˠl$PVMEZPVDIBOHFUIFTFUSBWFMMFS`TDIFDLT
ɹJOUPDBTI z ٙจ ˠl1MFBTFDBTIUIFTFUSBWFMMFS`TDIFDLTz ໋ྩจ ༁Ͱنଇ͚ͩͰॆͰ͋Γ
੍ݶର༁ ˗ ੍ݶݴޠ ༻ޠኮจ๏Λ੍ݶͯ͠จষΛهड़ ˠߏจߏҙຯղऍͷᐆດੑΛݮগͤ͞Δ ɹ͜ͱ͕ΒΕ͍ͯΔ ˗ ੍ݶର༁ ର༁ίʔύεʹ੍ݶݴޠͷߟ͑ํΛద༻
੍ݶର༁ ˗ ༁ੑ লུɾޠ͕গͳ͍ ˗ จ຺ࣗ༝ੑ ݪจͷ୯ޠྻͱ༁จͷ୯ޠྻ͕จ຺ใ ͳ͠ʹରԠ ˗ ޠॱҰகੑ
ݪจͷޠॱ͕༁จͰอͨΕ͍ͯΔ ˗ ༁ޠҰఆੑ ͋Δ୯ޠ͕ৗʹಉ͡୯ޠ༁͞ΕΔ
༁ੑείΞ ˗ 5Tର༁ࣙॻʹଘࡏ͢Δݪݴޠͷ୯ޠ ˗ 5Uର༁ࣙॻʹଘࡏ͢Δతݴޠͷ୯ޠ ˗ -ݪݴޠͱతݴޠͷ୯ޠͷ୯ޠϦϯΫ ˗ 5$3ର༁ରԠ
༁ੑείΞ ˗ ؙғΈ୯ޠର༁ࣙॻʹࡌ͍ͬͯΔ୯ޠ ˗ ܗଶૉղੳ݁Ռͱର༁ࣙॻͷΈͰஅ
ର༁ίʔύεϑΟϧλϦϯά ˗ ᮢʹΑΔϑΟϧλϦϯά ߴ༁ର༁ͷΈ͕Δ ෦ίʔύεαΠζ͕ݮগˠཏͰ͖ͳ͍ ˗ άϧʔϓ࠷ߴʹΑΔϑΟϧλϦϯά ݪݴޠ͕ಉ͡ର༁ͰάϧʔϓΛ࡞͠ɺ άϧʔϓͷ࠷ߴείΞΛ࣋ͭͷΛબ ˠ༁ର༁͕Δ
༁۟ͱҙ༁෦ͷׂߏங ˗ ର༁ରԠ5$3Λจ͚ͩͰͳ۟͘ʹద༻ ˗ จͱͯ͠ͷ5$3͍͕۟ͱͯ͠ߴ͍ ͷΛ༻ ˠཏͷԼΛ͑Δ ˗ ༁ର༁தʹ׳༻දݱͷΑ͏ͳ༁Ͱ ༁͖͢Ͱͳ͍ͷ͕ଘࡏ
༁۟ͱҙ༁෦ͷׂߏங ˗ ߴ༁۟ͦΕͱͦͷԼҐߏΛ൚Խͯ͠ ༁ࣝΛߏங ˗ ༁෦۟ରԠΛແࢹͨ͠༁ࣝΛߏங
࣮ݧ ˗ ӳػց༁ͷ༁࣭ͰධՁ ˗ #5&$ίʔύεͷ͏ͪ ର༁Λ༻ ཱྀߦձʹසग़͢ΔදݱΛूΊͨͷ ͷӳޠจ͕ෳͷຊޠ༁Λ࣋ͭ
ର༁ࣙॻߏஙํ๏ ˗ ࣙॻ" ίʔύεதͷڞىස͕ճҎ্ͷ୯ޠ ౷ܭతʹ୯ޠΞϥΠϝϯτΛߦ͏ ͦΕҎ֎γιʔϥεΛࢀর ద߹ɿɹ࠶ݱ ˗ ࣙॻ# ୯ޠΞϥΠϝϯτͷΈ
ద߹ɿɹ࠶ݱ
༁࣭ධՁํ๏ ˗ ࣗಈධՁ #-&6 HSBN Λ༻͍ͯධՁ ݪจ͋ͨΓͭͷࢀর༁ ˗ ओ؍ධՁ ςετηοτதͷจΛ͍ɺ
શର༁Λ༻ͨ͠ػց༁݁Ռͱൺֱͯ͠ ͲͪΒ͕Α͍༁͔Λఆ
ᮢʹΑΔίʔύεϑΟϧλϦϯάͷޮՌ ˗ ͍ͣΕ5$3㱢ͷͱ͖#-&6είΞ͕࠷େ ˗ ͞ΒʹᮢΛ্͛ΔͱϥϯμϜબʹൺͯ ༁࣭͕ٸܹʹѱԽ ˠ༁ੑ͍͕ඞཁͳର༁ ׳༻දݱͳͲ ͕ഉআ͞ΕͨͨΊ
ߏஙํ๏ͷҧ͍ͱ༁࣭ ˗ ϑΟϧλϦϯάʹΑΔख๏Ͱʙఔͷ ࣭վળ͔͠Ͱ͖ͳ͔ͬͨ ˠ༁ର༁ͷഉআͱΧόʔҡ࣋ͷཱ྆ࠔ ˗ ׂߏங#-&6είΞɺओ؍࣭ͱʹ্ ˠҙ༁ͷద༻݅Λݫ͘͠Ͱ͖ΔͨΊ
͓ΘΓʹ ˗ ༁ͷଟ༷ੑΛ༁ੑʹண੍ͯ͠ݶ ˗ ର༁ରԠʹΑΔϑΟϧλϦϯάͰ༁࣭Λ एׯ্ͤ͞Δ͜ͱ͕Ͱ͖Δ ˗ ର༁จΛ༁෦ͱҙ༁෦ʹׂ͠ɺ ҙ༁ͷద༻݅Λݫ͘͢͠Δ͜ͱͰ༁࣭Λ ্ͤ͞Δ͜ͱ͕Ͱ͖Δ
˗ ༁ޠҰٛੑจ຺ࣗ༝ੑΈࠐΉͱ༁࣭Λ ͞Βʹ্Ͱ͖Δͱਪଌ͞ΕΔ