Upgrade to PRO for Only $50/Year—Limited-Time Offer! 🔥
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
DeepNLP_BackPropagation_Rnn_and_Cnn
Search
izuna385
July 02, 2018
Science
0
150
DeepNLP_BackPropagation_Rnn_and_Cnn
深層学習による自然言語処理 2.5から2.9まで
izuna385
July 02, 2018
Tweet
Share
More Decks by izuna385
See All by izuna385
jel: japanese entity linker
izuna385
0
320
Firebase-React-App
izuna385
0
230
React+FastAPIを用いた簡単なWebアプリ作製
izuna385
0
1.5k
UseCase of Entity Linking
izuna385
0
510
Unofficial slides: From Zero to Hero: Human-In-The-Loop Entity Linking in Low Resource Domains (ACL 2020)
izuna385
1
630
Poly-encoders: Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring
izuna385
0
760
Zero-shot Entity Linking with Dense Entity Retrieval (Unofficial slides) and Entity Linking future directions
izuna385
3
970
Entity representation with relational attention
izuna385
0
74
Zero-Shot Entity Linking by Reading Entity Descriptions
izuna385
0
510
Other Decks in Science
See All in Science
マテリアルズ・インフォマティクスの先端で起きていること / What's Happening at the Cutting Edge of Materials Informatics
snhryt
1
140
機械学習を支える連続最適化
nearme_tech
PRO
1
170
butterfly_effect/butterfly_effect_in-house
florets1
1
110
240510 COGNAC LabChat
kazh
0
150
深層学習を利用して 大豆の外部欠陥を判別した研究事例の紹介
kentaitakura
0
240
Analysis-Ready Cloud-Optimized Data for your community and the entire world with Pangeo-Forge
jbusecke
0
110
証明支援系LEANに入門しよう
unaoya
0
400
トラブルがあったコンペに学ぶデータ分析
tereka114
2
1.1k
Machine Learning for Materials (Lecture 9)
aronwalsh
0
230
Factorized Diffusion: Perceptual Illusions by Noise Decomposition
tomoaki0705
0
260
Healthcare Innovation through Business Entrepreneurship
clintwinters
0
100
科学で迫る勝敗の法則(名城大学公開講座.2024年10月) / The principle of victory discovered by science (Open lecture in Meijo Univ. 2024)
konakalab
0
220
Featured
See All Featured
Building Flexible Design Systems
yeseniaperezcruz
327
38k
Building Better People: How to give real-time feedback that sticks.
wjessup
365
19k
RailsConf & Balkan Ruby 2019: The Past, Present, and Future of Rails at GitHub
eileencodes
132
33k
The Art of Programming - Codeland 2020
erikaheidi
53
13k
Fontdeck: Realign not Redesign
paulrobertlloyd
82
5.3k
Speed Design
sergeychernyshev
25
650
Building a Modern Day E-commerce SEO Strategy
aleyda
38
7k
The Invisible Side of Design
smashingmag
298
50k
The Pragmatic Product Professional
lauravandoore
32
6.3k
Fight the Zombie Pattern Library - RWD Summit 2016
marcelosomers
232
17k
A Philosophy of Restraint
colly
203
16k
Sharpening the Axe: The Primacy of Toolmaking
bcantrill
38
1.8k
Transcript
5 . 1 2
• : D !(#) 1 ∇!(#) • L 1 )
) ( 1 S G G 1 & '()*+, = .(/ 012 .(⋯ .(/ 2 '()*+, + 5(2)))) 62 67 68 69 /(:) ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ;2 ;7 /(:<2) = − 1 = = + 1
. . ! ℎ($) ℎ(&) ℎ(') (($) ((&) ((') ℎ(')=
( ' ( & ( $ ! *+(,) *- = *+(,) *.(/) 0 *.(/) *.(1) 0 *.(1) *- •
. . ! ℎ($) ℎ(&) '($) '(&)
. .
None
(
.( )
( )
2'+ )NN 2 #) 0%&1&-" (3*, )
.( !/$→ → Residual Connection, Batch Nomalization( ) ! Loss func ! Loss func
: Residual Connection –– F(x) (→-!()) F(x) + x
→ & " - Identity Mapping ' : -*&%,+' ./$ # Identity – [1] He, Kaiming, et al. "Identity mappings in deep residual networks." European Conference on Computer Vision. Springer, Cham, 2016. . .
. . (2.33) (2.34)
0 1 . 2 C2 2 " 3 2 3
2 ) 2 2 ( 3 23 2 !"#$ !" %"&$ %"#$ %" %"&$ '() '() '() '*+, '*+, -"#$ = /(!"#$ ) -" -"&$ %" !" M I NR 2 '*+, O L '() : input P / L !" = '*+, 2 !"#$ + '() %" ,, L
2 . 3 !" #$% &% !" #$' !" #$(
!% #$% !% #$' !% #$( !' #$% !' #$' !' #$( !( #$% !( #$' !( #$( &' &( )% )' )(
2 . 3 !" #$% &% !" #$' !" #$(
!% #$% !% #$' !% #$( !' #$% !' #$' !' #$( !( #$% !( #$' !( #$( &' &( )% )' )(
2 . 3 !" #$% &% !" #$' !" #$(
!% #$% !% #$' !% #$( !' #$% !' #$' !' #$( !( #$% !( #$' !( #$( &' &( )% )' )(
2 . 3 !" #$% &% !" #$' !" #$(
!% #$% !% #$' !% #$( !' #$% !' #$' !' #$( !( #$% !( #$' !( #$( &' &( )% )' )( • 23 !* # 1
: !"#$ !%& '( )( … … )*
'* ………… ………… +( +* !"#$ (-) !%& (-) '% …… '/ )/ +/
. - •
) (
8 99 2 :9.8 9 9 6 5 3 2
2 5 2 28 79 8 3 9 1 56 2 59 7 /0-
-5 1 02 1 25 58 8 ./ 8 .2
0 8
22/1 444 1 1 - 2 3--.-.3-. 11
http://deeplearning.stanford.edu/wiki/index .php/Feature_extraction_using_convolution
/885 6 0: 6. 5 5
RNN Vanishing/Exploding Gradient : !"#$ !%&
'( )( … … )* '* ………… ………… +( +* !"#$ (-) !%& (-) '% …… '/ )/ +/