Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
[Paper Reading] Attention is All You Need
Search
Daiki Tanaka
August 04, 2019
Research
140
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
[Paper Reading] Attention is All You Need
presentation slides for "Attention is All You Need" (
https://arxiv.org/abs/1706.03762
)
Daiki Tanaka
August 04, 2019
More Decks by Daiki Tanaka
See All by Daiki Tanaka
カーネル法概観
daikitanak
0
770
Counterfactual learning to rank: introduction
daikitanak
0
910
カーネル法:正定値カーネルの理論
daikitanak
0
90
[Paper reading] L-SHAPLEY AND C-SHAPLEY: EFFICIENT MODEL INTERPRETATION FOR STRUCTURED DATA
daikitanak
1
220
Interpretability of Machine Learning : Paper reading (LIME)
daikitanak
0
190
[Paper reading] Local Outlier Detection With Interpretation
daikitanak
0
95
Other Decks in Research
See All in Research
[UAI2026, Tutorial Talk] Tensor Networks for Density Estimation: The merits of low-rank structures, tractable inference, and uncertainty quantification
gkazunii
0
110
ふとした出会いで生まれたSkillが、 社内利用1位になるまで
mikimhk
22
25k
Pesée et pointage par imagerie 3D : PHENO3D entre en phase de test terrains
institutdelelevage
PRO
0
150
Détecter les vulnérabilités dans un environnement souverain
hellosct1
0
120
[WebDB2026]セレンディピティ指向推薦システム再考 ―セレンディピティの原義・類型・発生過程に基づく設計指針―
recsyslab
PRO
0
220
[最先端NLP勉強会2026] Checklists Are Better Than Reward Models For Aligning Language Models
nzw0301
1
340
SPACE 2026 - Renforcer l’attractivité des métiers des productions animales dans les établissements d'enseignement agricole
institutdelelevage
PRO
0
120
Du suivi individuel à la sélection animale : comment Digi4Live valorise les données des technologies numériques en élevage
institutdelelevage
PRO
0
110
Easy to Guess, Hard to Verify: Lessons from AIMO 3 for Olympiad-Level AI Mathematics
corochann
0
150
Adapter les bergeries aux fortes chaleurs
institutdelelevage
PRO
0
170
ros2-perf-multihost: 分散システムにおける客観的なアーキテクチャ評価フレームワーク
takasehideki
0
320
湯村研究室の紹介2026 / yumulab2026
yumulab
0
200
Featured
See All Featured
How to Ace a Technical Interview
jacobian
280
24k
A Tale of Four Properties
chriscoyier
163
24k
Writing Fast Ruby
sferik
630
63k
Navigating Algorithm Shifts & AI Overviews - #SMXNext
aleyda
1
1.7k
It's Worth the Effort
3n
188
29k
Are puppies a ranking factor?
jonoalderson
2
4.1k
AI: The stuff that nobody shows you
jnunemaker
PRO
10
1.2k
Sam Torres - BigQuery for SEOs
techseoconnect
PRO
0
600
The Illustrated Children's Guide to Kubernetes
chrisshort
51
53k
How to Get Subject Matter Experts Bought In and Actively Contributing to SEO & PR Initiatives.
livdayseo
0
240
Digital Ethics as a Driver of Design Innovation
axbom
PRO
1
490
Intergalactic Javascript Robots from Outer Space
tanoku
273
28k
Transcript
, , , , , , ,
• : ( 1 2 • C G • )I
G ( • )G G 7 GK B • 1 C B • 1 A 2 C B G : • . A 9 6G A B G : • 0B . G B G : • ) 1 C 6 J G L • NIA G G B • BB 2 B GIA
• . • • •
• . .. . . ,. . . • .
. . . . , . . . . . • . . . . . ., . . . ,. . • . ., . ., . . .,. . , . ., . . ., . .
• . , . . . . . , .
, . . • . . . . . , . . . . . . . . . .
• , , , .. . . , ,
, • , - , , . , . , , . , .,
• . • . . . ("# , "% ,
… , "' ) • . • • . . . ()# , )% , … , )' ) • :
• • ( • ) • - - • -
• - - - • ( ) ) ) :) )
-- - -- - -- -
• ,, ), ) , ( , ( , →
, "##$%#&'% ()$*+, -$+, ./0)$ = 2'3#4/5 ()$*+ 6 -$+7 6 ./0)$
• !""#$"%&$ '(#)*, ,#*, -./(# = 1&2"3.4 '(#)* 5 ,#*6
5 -./(# . . , . . . . , . .
• ) ( ( ( ( , , → (
"##$%#&'% ()$*+, -$+, ./0)$ = 2'3#4/5 ()$*+ 6 -$+7 6 ./0)$ • ( , ( ( ( ( ,( , , , →, "##$%#&'% 8)$*+, -$+, ./0)$ = 2'3#4/5 8)$*+ 6 -$+7 6 ./0)$
- - -- - • . .. - • !"#$%&'()(*
+, - + /)) • !"#$%&' +-1 • . • . 2$$34$5"4 +, -, 6 = !"#$%&' +-1 89 6 89 - . . - . --
• . - , , . - 8 - ,
. . , , . 8 . . . 8 . • ) . . 8 ., ( - . = - . . 8 ., !", !" $%! !& - . ) . . !" = !& = ()*+,- . ) 0123456$! 7, 8, 9 = :;%<$3 ℎ6$!>, … , ℎ6$!. @A Bℎ6C6 ℎ6$!D = E336%34;%(7@ D G, 8@D H, 9@D I)
-- • , : 3 2 A 2 A 2
1 2: : : 22 2: 12 12 12 2: : , 2 A2 2 2 3 2 :1 2 ) :1 A2 - 2 3 2 1 : 2: : 2 : 2 3 2: : 2 : 2 2: 12 ( 3 2 2 ) A2 - :1 A2 2 2 3 2 2 2 : 2 2 2 3 2: : 2 : 2 12 12 ) - 2 3 2 .2 :221 2 2: 23 1 :3 : 3 : 2 12 12 2 2 2 2 A 2 2 2 2
- ) - ) ( . - ..
-- / .
• . ) . , . ) . ( -
, - . , ( , . , . , ) ) • , ) )) . - - . ) , , )) )) ( ) ( ) . -
) 3 , 4 2 3 , 4 3. 2
. 3 . )1. 4 , 4 2 1 , . 33.32 .- )1. 1 3. 1 . .. 3 . -. . -. ,2. ( 2-. . . 2 .3 . 2 , 3- 2.3- 4 . 2 . . 3. 4 -.3
None
- - - • . . . . -- -
- . !!" # = %&'( #)* + ,* )- + ,- 4 4 4 4 4 4 4 3 1 ( ) 2 %&'( #* )* + ,* )- + ,- %&'( #- )* + ,* )- + ,- %&'( #. )* + ,* )- + ,- %&'( #/ )* + ,* )- + ,- %&'( #0 )* + ,* )- + ,- %&'( #1 )* + ,* )- + ,- %&'( #2 )* + ,* )- + ,- 2 3 1 3 1 3 1 3 1 3 1 3 1
None
• , . • , • , !" #$%, 2(
= sin #$% 10000 /0 123456 , !" #$%, 2( + 1 = cos #$% 10000 /0 123456 • #$% , , ( , :
.0 +- +- 1 - + -1+.- +- +- .0
/ . / . / / : / / / /
. - .. -- / .
.
. - .. -- / .
- • ., ,1, 2 ,1 :, 0, ., :
: . ., . ., 1 . : : ., , : , . 0 - • ., : ,1, 2 ,1 . ., .,
• • 5 ), . 5 , 5
5 ) . . =) . ) • ) 5 . ) ( 5 5 ) ) ) . ( 5 5 5 • 5 . , , ) ) ) )
- - - • • 5: ( 0GBED 2A F
G , ( ) FDEEDHG AG AG A I D • 5: ( 0GBED 1 AG , FDEEDHG AG AG A • 3 N A G A MEA • 6 4 4- 8 28 • . A FH AE , AI H HM • .DB FH AE , AI ) O • I DFDPA , - F • F MI G AG A A A EA GDGB A
• . ( . .= <= • ( . .=
= < 4 . 4 = . < .= , . = =4 ), = . < .= . < . !"#$ = !&'()* + exp / 012 3 1 5 67890 < . • !&'()* : . = = : 4 , ), • 90 = ∑; <=> 0?@A>B CD 0EFBG@H HG@> I0 JK; G0L MK; ∑; <=> 0?@A>B CD 0EFBG@H I0 JK;
None
None
- -- - -
- -- - - . .
• . . - . . . - ,
. , . . . , . . , - -- ., . . - , , . . • . - . . . . , . . . . , . , ,
• -- / . / - • . . .
/ /- . . / -