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
Problems of Neural Networks and its solutions
Search
izuna385
June 21, 2018
Technology
0
130
Problems of Neural Networks and its solutions
Residual Connections とBatch Normalizationがメイン
izuna385
June 21, 2018
Tweet
Share
More Decks by izuna385
See All by izuna385
jel: japanese entity linker
izuna385
0
340
Firebase-React-App
izuna385
0
240
React+FastAPIを用いた簡単なWebアプリ作製
izuna385
0
1.6k
UseCase of Entity Linking
izuna385
0
550
Unofficial slides: From Zero to Hero: Human-In-The-Loop Entity Linking in Low Resource Domains (ACL 2020)
izuna385
1
650
Poly-encoders: Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring
izuna385
0
840
Zero-shot Entity Linking with Dense Entity Retrieval (Unofficial slides) and Entity Linking future directions
izuna385
3
1k
Entity representation with relational attention
izuna385
0
80
Zero-Shot Entity Linking by Reading Entity Descriptions
izuna385
0
540
Other Decks in Technology
See All in Technology
AWSLambdaMCPServerを使ってツールとMCPサーバを分離する
tkikuchi
1
3k
Mastraに入門してみた ~AWS CDKを添えて~
tsukuboshi
0
270
AIエージェント開発手法と業務導入のプラクティス
ykosaka
2
1.4k
【Λ(らむだ)】最近のアプデ情報 / RPALT20250422
lambda
0
110
Goの組織でバックエンドTypeScriptを採用してどうだったか / How was adopting backend TypeScript in a Golang company
kaminashi
6
6.2k
DuckDB MCPサーバーを使ってAWSコストを分析させてみた / AWS cost analysis with DuckDB MCP server
masahirokawahara
0
1.3k
彩の国で始めよう。おっさんエンジニアから共有したい、当たり前のことを当たり前にする技術
otsuki
0
150
Porting PicoRuby to Another Microcontroller: ESP32
yuuu
4
430
4/16/25 - SFJug - Java meets AI: Build LLM-Powered Apps with LangChain4j
edeandrea
PRO
2
120
読んで学ぶ Amplify Gen2 / Amplify と CDK の関係を紐解く #jawsug_tokyo
tacck
PRO
1
160
Amazon CloudWatch を使って NW 監視を行うには
o11yfes2023
0
170
LLM as プロダクト開発のパワードスーツ
layerx
PRO
1
240
Featured
See All Featured
BBQ
matthewcrist
88
9.6k
Java REST API Framework Comparison - PWX 2021
mraible
30
8.5k
Bash Introduction
62gerente
611
210k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
507
140k
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
9
760
Documentation Writing (for coders)
carmenintech
69
4.7k
CoffeeScript is Beautiful & I Never Want to Write Plain JavaScript Again
sstephenson
160
15k
Building a Modern Day E-commerce SEO Strategy
aleyda
40
7.2k
Raft: Consensus for Rubyists
vanstee
137
6.9k
How STYLIGHT went responsive
nonsquared
99
5.5k
[RailsConf 2023 Opening Keynote] The Magic of Rails
eileencodes
29
9.4k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
32
2.2k
Transcript
1 / 18 Neural Networks
2 / 18 1. NN !
• Residual Network • Batch Normalization 2. 1. • •
3 / 18 Plain NNs(&) ' pros #%
" (ex. CNN, RNN, ...) cons ! $ $
4 / 18 RNN RNN [1] P. Razvan et
al ,"On the difficulty of training recurrent neural networks." International Conference on Machine Learning. 2013. !"#$ !" %"&$ %"#$ %" %"&$ '() '() '() '*+, '*+, -!"# = /(!!"# ) -! -!$# %! : input !! : hidden state '%&' : '() : input / !" = '*+, 2 !"#$ + '() %"
5 / 18 !" !# !$ %" %# %$ &'(
&'( &'( &)*+ &)*+ ,! = .(!! ) ," ,# RNN 3 1, 12 = 1," 12 + 1,# 12 + 1,$ 12 1,$ 12 = 4 "565$ 1,$ 1!$ 7 1!$ 1!6 7 18!6 12 1!$ 1!" = 1!$ 1!# 7 1!# 1!" = &)*+ 9 :;<= >? !# 7 &)*+ 9 :;<= >? !" @A!B @C : !" ~!6E" fix !6
6 / 18 RNN Vanishing/Exploding Gradient : !"#$ !%&
'( )( … … )* '* ………… ………… +( +* !"#$ (-) !%& (-) '% …… '/ )/ +/
7 / 18 ,$+ /' !"#$ !- !"#$ 2 %
× '()* + ×%,- → # !"#$ !"#$ . 2 % × '()*(+).,-×%,- 1%input or 1)* Loss( RNN ."0& Vanishing/Exploding Gradient
8 / 18 +$ DeepNN( ! +
" )*&!/#% ' (→ ! Loss func ! Loss func → Residual Connection, Batch No malization
9 / 18 0), : Residual Connection – -– F(x)
"/#2 → "/ F(x) + x → (4 '$"/ Identity Mapping +%*1&: 3 . ! 3 Identity – [1] He, Kaiming, et al. "Identity mappings in deep residual networks." European Conference on Computer Vision. Springer, Cham, 2016.
10 / 18 : Residual Connection –– ' Forward
$#& Backward !$"& Deep % & input
11 / 18 Residual Connection –– https://icml.cc/2016/tutorials/icml2016_tutorial_deep_residual_networks_kaiminghe.pdf
12 / 18 ResNet Batch Normalization ResNet Residual Block
• ImplementationBatch Normalization NN ! $# • Batch Normalization" ## http://torch.ch/blog/2016/02/04/resnets.html Plain
13 / 18 ( ) 1 2
( ) n … Batch Normalization –Revisit Gaussian-
14 / 18 Batch Normalization -Input Data distribution
- (Convergence) !! Input NN → input
15 / 18 Batch Normalization -distribution - !"#$% & '
= ) & ' ← ' − , - ~/(,, -2) input
16 / 18 Batch Normalization Data distribution •
=(!, ")fix • Batch Normalization Batch Normalization
17 / 18 Batch Normalization – [2]Ioffe, Sergey,
and Christian Szegedy. "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift." (2015). !, # !%$( → normalize scaling '"&# nomalize
18 / 18 DeepNN+ ! /
& -"#.#)%/'( *$ +!→ , Identity – normalize scaling implement Deep Net