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
DeepNLP_BackPropagation_Rnn_and_Cnn
Search
izuna385
July 02, 2018
Science
190
0
Share
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
More Decks by izuna385
See All by izuna385
jel: japanese entity linker
izuna385
0
470
Firebase-React-App
izuna385
0
270
React+FastAPIを用いた簡単なWebアプリ作製
izuna385
0
1.8k
UseCase of Entity Linking
izuna385
0
650
Unofficial slides: From Zero to Hero: Human-In-The-Loop Entity Linking in Low Resource Domains (ACL 2020)
izuna385
1
710
Poly-encoders: Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring
izuna385
0
960
Zero-shot Entity Linking with Dense Entity Retrieval (Unofficial slides) and Entity Linking future directions
izuna385
3
1.2k
Entity representation with relational attention
izuna385
0
110
Zero-Shot Entity Linking by Reading Entity Descriptions
izuna385
0
620
Other Decks in Science
See All in Science
KISHIMOTO Atsuo
genomethica
0
140
Bリーグのショットデータを活用した得点期待値モデルの構築 / Construction of expected points model using shot data of B.LEAGUE
konakalab
0
130
People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text
rudorudo11
0
250
ITTF卓球世界ランキングのポイント比を用いた試合結果予測モデルの性能評価 / Performance evaluation of match result prediction models using the point ratio of the ITTF Table Tennis World Ranking
konakalab
0
130
次代のデータサイエンティストへ~スキルチェックリスト、タスクリスト更新~
datascientistsociety
PRO
3
42k
HajimetenoLT vol.17
hashimoto_kei
1
230
ハミルトン・ヤコビ方程式の解の性質と物理的意味
enakai00
0
530
Endel Tulvingとエピソード記憶
rmaruy
0
130
生成AIの現状と展望
tagtag
PRO
0
130
力学系から見た現代的な機械学習
hanbao
4
4.2k
データベース06: SQL (3/3) 副問い合わせ
trycycle
PRO
1
900
NDCG is NOT All I Need
statditto
2
3.1k
Featured
See All Featured
What’s in a name? Adding method to the madness
productmarketing
PRO
24
4.1k
Imperfection Machines: The Place of Print at Facebook
scottboms
270
14k
Making Projects Easy
brettharned
120
6.7k
Stop Working from a Prison Cell
hatefulcrawdad
274
21k
Unlocking the hidden potential of vector embeddings in international SEO
frankvandijk
0
830
The Pragmatic Product Professional
lauravandoore
37
7.3k
The AI Search Optimization Roadmap by Aleyda Solis
aleyda
1
5.8k
GitHub's CSS Performance
jonrohan
1033
470k
Public Speaking Without Barfing On Your Shoes - THAT 2023
reverentgeek
1
410
How to Get Subject Matter Experts Bought In and Actively Contributing to SEO & PR Initiatives.
livdayseo
0
130
Thoughts on Productivity
jonyablonski
76
5.2k
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
133
19k
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 : !"#$ !%&
'( )( … … )* '* ………… ………… +( +* !"#$ (-) !%& (-) '% …… '/ )/ +/