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
VAE
Search
Atom
February 21, 2019
0
160
VAE
2020/3/15: youtubeのコメントによるご指摘を頂き,6ページの図を差し替えました
Atom
February 21, 2019
Tweet
Share
More Decks by Atom
See All by Atom
文献紹介 / Structure-based Knowledge Tracing: An Influence Propagation View
roraidolaurent
0
88
文献紹介 / Knowledge Tracing with GNN
roraidolaurent
0
92
文献紹介 / Non-Intrusive Parametric Reduced Order Models withHigh-Dimensional Inputs via Gradient-Free Active Subspace
roraidolaurent
0
55
ニューラルネットワークのベイズ推論 / Bayesian inference of neural networks
roraidolaurent
1
2.8k
Graph Convolutional Networks
roraidolaurent
0
230
文献紹介 / A Probabilistic Annotation Model for Crowdsourcing Coreference
roraidolaurent
0
65
文献紹介Deep Temporal-Recurrent-Replicated-Softmax for Topical Trends over Time
roraidolaurent
0
96
文献紹介/ Bayesian Learning for Neural Dependency Parsing
roraidolaurent
0
120
ポッキー数列の加法定理 / Pocky number additon theorem
roraidolaurent
0
210
Featured
See All Featured
Design and Strategy: How to Deal with People Who Don’t "Get" Design
morganepeng
130
19k
A Tale of Four Properties
chriscoyier
160
23k
I Don’t Have Time: Getting Over the Fear to Launch Your Podcast
jcasabona
32
2.3k
The Illustrated Children's Guide to Kubernetes
chrisshort
48
50k
Building a Scalable Design System with Sketch
lauravandoore
462
33k
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
31
2.4k
Art, The Web, and Tiny UX
lynnandtonic
299
21k
Scaling GitHub
holman
459
140k
Writing Fast Ruby
sferik
628
62k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
29
1.8k
How STYLIGHT went responsive
nonsquared
100
5.6k
Testing 201, or: Great Expectations
jmmastey
42
7.5k
Transcript
AEとVAE 変分オートエンコーダーとは 第7回 B3勉強会 2019/2/21 長岡技術科学大学 自然言語処理研究室 吉澤 亜斗武
参考文献・資料 [1] IIBMP2016:深層生成モデルによる表現学習 https://www.slideshare.net/pfi/iibmp2016-okanohara-deep-generative-models-for-representation- learning [2] @kenchin110100:AutoEncoder, VAE, CVAEの比較 〜なぜVAEは
連続的な画像を生成できるのか?〜 https://qiita.com/kenchin110100/items/7ceb5b8e8b21c551d69a [3] 渡辺澄夫:オートエンコーダー http://watanabe- www.math.dis.titech.ac.jp/users/swatanab/Renshu_3.pdf 2
(1) VAEとは ・Auto-encoder 教師なし学習の一つ 識別モデル(Discriminative model)のニューラルネット 入力を受けて出力が決定論的に決まる 非線形の次元圧縮が可能(線形は主成分解析) 3
引用:[3] 4
(1) VAEとは ・Variational Auto-encoder 教師なし学習の一つ 生成モデル(Generrative model)のニューラルネット 入力を受けて出力が確率的に決まる ELBOを最大化 5
引用[1]
(1) VAEとは 6 引用[2]
(1) VAEとは 引用[1] 7