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
EMNLP2015読み会:Effective Approaches to Attention-...
Search
tkng
October 24, 2015
Research
2
4k
EMNLP2015読み会:Effective Approaches to Attention-based Neural Machine Translation
tkng
October 24, 2015
Tweet
Share
More Decks by tkng
See All by tkng
LSTMを用いた自然言語処理について
tkng
3
3.7k
自然言語処理と深層学習の最先端
tkng
16
7.7k
basis-of-optimization.pdf
tkng
1
1.4k
Other Decks in Research
See All in Research
cvpaper.challenge 10年の軌跡 / cvpaper.challenge a decade-long journey
gatheluck
3
300
能動適応的実験計画
masakat0
2
780
[CV勉強会@関東 CVPR2025] VLM自動運転model S4-Driver
shinkyoto
2
460
多言語カスタマーインタビューの“壁”を越える~PMと生成AIの共創~ 株式会社ジグザグ 松野 亘
watarumatsuno
0
110
Agentic AIとMCPを利用したサービス作成入門
mickey_kubo
0
430
Combinatorial Search with Generators
kei18
0
640
Cross-Media Information Spaces and Architectures
signer
PRO
0
230
A scalable, annual aboveground biomass product for monitoring carbon impacts of ecosystem restoration projects
satai
3
160
20250725-bet-ai-day
cipepser
2
390
経済学と機械学習:因果推論と密度比推定を中心に
masakat0
0
140
最適化と機械学習による問題解決
mickey_kubo
0
160
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
shunk031
14
9.7k
Featured
See All Featured
We Have a Design System, Now What?
morganepeng
53
7.7k
The Invisible Side of Design
smashingmag
301
51k
The Art of Programming - Codeland 2020
erikaheidi
55
13k
Art, The Web, and Tiny UX
lynnandtonic
302
21k
The Straight Up "How To Draw Better" Workshop
denniskardys
236
140k
GraphQLとの向き合い方2022年版
quramy
49
14k
Six Lessons from altMBA
skipperchong
28
4k
Rails Girls Zürich Keynote
gr2m
95
14k
How to Ace a Technical Interview
jacobian
279
23k
Designing for Performance
lara
610
69k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
48
9.6k
BBQ
matthewcrist
89
9.8k
Transcript
Effective Approaches to Attention-based Neural Machine Translation Authors: Minh-Thang LuongɹHieu
PhamɹChristopher D. Manning ಡΉਓ: ಙӬ೭ ਤશͯ͜ͷจ͔ΒҾ༻ &./-1ಡΈձ
ࣗݾհɿಙӬ೭ • Twitter ID: @tkng • εϚʔτχϡʔεגࣜձࣾͰNLPͬͯ·͢
ࠓͷจʁ • Effective Approaches to Attention-based Neural Machine Translation •
ڈ͙Β͍͔ΒྲྀߦΓ࢝Ίͨseq2seqܥͷख ๏ͷ֦ு
Seq2seq modelͱʁ • Encoder/Decoder modelͱݴ͏ • ༁ݩͷจΛݻఆͷϕΫτϧʹΤϯίʔυ ͯ͠ɺ͔ͦ͜Β༁ޙͷจΛσίʔυ͢Δ • ՄมͷσʔλऔΓѻ͍͕͍͠ͷͰɺ
͑ͯݻఆʹͯ͠͠·͏ͱ͍͏ൃ
Ͳ͏ͬͯݻఆʹΤϯίʔυ ͢Δͷʁ • recurrent neural networkΛ͏ • http://colah.github.io/posts/2015-08-Understanding-LSTMs/ • http://kaishengtai.github.io/static/slides/treelstm-acl2015.pdf
• LSTM = recurrent neural networkͷҰछ
Seq2seqϞσϧͰͷ༁
Seq2seq·ͰͷಓͷΓ (1) • Recurrent Continuous Translation Models (EMNLP2013) • Learning
Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation (EMNLP2014)
Seq2seq·ͰͷಓͷΓ (2) • Sequence to Sequence Learning with Neural Networks
(NIPS2014) • ൺֱతγϯϓϧͳStacked LSTM͕ྑ͍ੑೳΛ ࣔ͢͜ͱ͕࣮ݧͰࣔ͞Εͨ • ϏʔϜαʔνɺٯॱͰͷೖྗɺΞϯαϯϒϧ ͷ3छྨͷ͕ೖ͍ͬͯΔ
Seq2seqϞσϧͷऑ • จʹऑ͍ • ݻ༗໊ࢺ͕ೖΕସΘΔ
AttentionʹΑΔվળ [Bahdanau+ 2015] • DecodeͷࡍͷContextʹEncodeͷࡍͷ֤࣌ࠁ ʹ͓͚ΔӅΕঢ়ଶͷॏΈ͖Λ༻͍Δ • ॏΈࣗମRNNͰܭࢉ͢Δ
ࠓճͷจͷߩݙ • ৽͍͠attention (local attention) ΛఏҊͨ͠ • ༁ݩจʹ͓͍ͯɺҐஔɹ͔ΒલޙD୯ޠ ͷӅΕঢ়ଶͷॏΈ͖ΛऔΔ •
ॏΈͷܭࢉglobal attentionͷ߹ͱಉ༷ • ɹ1ͭͣͭਐΊ͍ͯ͘߹ʢlocal-mʣ ͱɺ͜ΕࣗମRNNʹ͢Δ߹ʢlocal- pʣͷ2ͭΛ࣮ݧ͍ͯ͠Δ pt pt
local attention
local attentionͷҹ • ޠॱ͕ࣅ͍ͯΔݴޠؒͰͷ༁ͳΒɺ໌Β͔ ʹ͜ͷํ͕ྑͦ͞͏ • ӳΈ͍ͨʹޠॱ͕େ͖͘ҧ͏߹ɺ Ґஔɹͷਪఆࣗମ͕͍͠λεΫʹͳͬͪΌ ͍ͦ͏… pt
࣮ݧ݁ՌɿWMT'14
࣮ݧ݁ՌɿWMT'14 • Α͘ݟΔͱɺlocal attentionͰͷੑೳ্ +0.9ϙΠϯτ • ଞͷςΫχοΫͰՔ͍ͰΔϙΠϯτ͕ଟ͍
࣮ݧ݁ՌɿWMT'15
͍͔ͭ͘༁αϯϓϧ
·ͱΊ • Seq2seqϞσϧͷ֦ுͱͯ͠ɺlocal attention ΛఏҊͨ͠ • ఏҊख๏͍͔ͭ͘ͷ࣮ݧʹ͓͍ͯɺState of the artͷੑೳΛୡͨ͠
ײ • Local attentionΛඍ • ྨࣅ͢Δख๏ͱ۩ମతʹͲ͏ҧ͏͔͕໌շʹ ॻ͔Ε͓ͯΓɺಡΈ͔ͬͨ͢ • AttentionΛཧղͰ͖ͯΑ͔ͬͨʢখฒײʣ