SlideShare ist ein Scribd-Unternehmen logo
1 von 21
1
DEEP LEARNING JP
[DL Papers]
http://deeplearning.jp/
“SimGANs: Simulator-Based Generative Adversarial Networks
for ECG Synthesis to Improve Deep ECG Classification (ICML2020)”
Naoki Nonaka
2020/8/28
2020/8/28 2
目次
• 書誌情報
• 背景
• 提案手法
• 実験
• まとめ
2020/8/28 3
書誌情報
• 会議:ICML 2020
• 著者:Tomer Golany, et al.
(Israel Institute of Technology)
• 実装:https://github.com/tomerGolany/sim_gan
2020/8/28 4
概要
GANによる心電図の生成に取り組んだ研究
心電図のシミュレータを用いて,生成対象に関する情報を
与えることで生成の質を向上するSimGANを提案
提案手法による生成データを用いることで,
心電図の分類精度を向上できることを示した
2020/8/28 5
背景
 (生成モデルにより)生成したデータで分類精度を向上したい
 心電図では,データの生成過程に関する知識に基づくシミュレータが存在
常微分方程式(ODE)によるシミュレータを活用した生成モデル
2020/8/28 6
心電図とは
基本的な心臓の検査の一つ
不整脈や心筋梗塞,狭心症を調べる上で重要
心臓の電気的活動を電極で検出して記録
https://physionet.org/content/challenge-2017/1.0.0/
収集されるデータの例
2020/8/28 7
提案手法: SimGAN
Discriminator:
Generator:
通常のGANで用いられるLoss シミュレータとの一致を評価するLoss
GANのGeneratorに,シミュレータとの一致を評価する損失を追加
2020/8/28 8
提案手法: SimGAN
: ℎ𝑖は時刻𝑖における値,ℎは固定長𝐿の心電図
: 𝑙 ~ 𝑙 + 1 までの(生成された)心電図の変化の割合
: シミュレータによる心電図の変化の割合
2020/8/28 9
心電図のシミュレータ
以下のODEで表現される(𝑥 𝑡 , 𝑦 𝑡 , 𝑧 𝑡 )の3つの軌跡で心電図を再現
(McSharry et al.; 2003 )
3次元の軌跡として表現される
心電図の波形として用いられるのはzのみ
2020/8/28 10
提案手法: SimGAN
Generatorの構造とGeneratorのLoss
2020/8/28 11
実験
 生成データによる分類精度の変化検証
 Generatorへの入力の比較
 シミュレータの出力を直接用いる場合との比較
 生成データの定性的評価
2020/8/28 12
データ
MIT-BIH arrhythmia database
 30minの心電図 x 48 (360Hzで収集)
 109,492心拍
 クラスラベル
 Ventricular Ectopic Beat (VEB) | 心室性期外収縮
 Supraventricular Ectopic Beat (SVEB) | 上室性期外収縮
 Fusion Beat | 融合心拍
 Normal Beat
2020/8/28 13
比較モデル: 実験1 (生成データによる分類精度の変化検証)
G
VGAN / DCGAN SimVGAN / SimDCGAN
G
Al Rahhal et al.
[ 追加データなし ]
2020/8/28 14
生成データによる分類精度の変化検証
𝐿 𝐺
𝐸𝑈𝐿
(𝜙 𝐺) を導入して生成したECGを用いた場合に精度が向上
2020/8/28 15
比較モデル: 実験2 (Generatorへの入力比較)
Refine GAN
G
SimVGAN / SimDCGAN
G
(Simulator出力)
2020/8/28 16
Generatorへの入力の比較
シミュレータの出力をRefinesした場合よりも精度が向上
2020/8/28 17
比較モデル: 実験3 (シミュレータの出力を直接用いる場合との比較)
ECG simulator SimVGAN / SimDCGAN
G
(Simulator出力)
[ - ]
[ - ]
[ - ]
2020/8/28 18
シミュレータの出力を直接用いる場合との比較
シミュレータの出力を直接用いる場合よりも精度が向上
2020/8/28 19
生成されたECGの定性的評価
通常のGANよりも生成データの質が高い
2020/8/28 20
まとめ
 シミュレータに関する知識を組み込んだSimGANを提案
 提案手法により生成した心電図を用いることで,
分類問題の精度が向上することを示した
 提案手法により生成された心電図が既存手法と比較して,
実際の心電図に近いことを定性的に確認した
2020/8/28 21
比較モデル
G
VGAN / DCGAN Refine GAN
G
SimVGAN / SimDCGAN
G
(Simulator出力)

Weitere ähnliche Inhalte

Mehr von Deep Learning JP

【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?Deep Learning JP
 
【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究についてDeep Learning JP
 
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )Deep Learning JP
 
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...Deep Learning JP
 
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"Deep Learning JP
 
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "Deep Learning JP
 
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat ModelsDeep Learning JP
 
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"Deep Learning JP
 
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...Deep Learning JP
 
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...Deep Learning JP
 
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...Deep Learning JP
 
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...Deep Learning JP
 
【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...
【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...
【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...Deep Learning JP
 
【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...
【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...
【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...Deep Learning JP
 
【DL輪読会】マルチモーダル 基盤モデル
【DL輪読会】マルチモーダル 基盤モデル【DL輪読会】マルチモーダル 基盤モデル
【DL輪読会】マルチモーダル 基盤モデルDeep Learning JP
 
【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...
【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...
【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...Deep Learning JP
 
【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...
【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...
【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...Deep Learning JP
 
【DL輪読会】大量API・ツールの扱いに特化したLLM
【DL輪読会】大量API・ツールの扱いに特化したLLM【DL輪読会】大量API・ツールの扱いに特化したLLM
【DL輪読会】大量API・ツールの扱いに特化したLLMDeep Learning JP
 
【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision
【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision
【DL輪読会】DINOv2: Learning Robust Visual Features without SupervisionDeep Learning JP
 
【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...
【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...
【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...Deep Learning JP
 

Mehr von Deep Learning JP (20)

【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?【DL輪読会】Can Neural Network Memorization Be Localized?
【DL輪読会】Can Neural Network Memorization Be Localized?
 
【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について【DL輪読会】Hopfield network 関連研究について
【DL輪読会】Hopfield network 関連研究について
 
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
【DL輪読会】SimPer: Simple self-supervised learning of periodic targets( ICLR 2023 )
 
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
【DL輪読会】RLCD: Reinforcement Learning from Contrast Distillation for Language M...
 
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
【DL輪読会】"Secrets of RLHF in Large Language Models Part I: PPO"
 
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "【DL輪読会】"Language Instructed Reinforcement Learning  for Human-AI Coordination "
【DL輪読会】"Language Instructed Reinforcement Learning for Human-AI Coordination "
 
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
【DL輪読会】Llama 2: Open Foundation and Fine-Tuned Chat Models
 
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
【DL輪読会】"Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware"
 
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
【DL輪読会】Parameter is Not All You Need:Starting from Non-Parametric Networks fo...
 
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
【DL輪読会】Drag Your GAN: Interactive Point-based Manipulation on the Generative ...
 
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
【DL輪読会】Self-Supervised Learning from Images with a Joint-Embedding Predictive...
 
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
【DL輪読会】Towards Understanding Ensemble, Knowledge Distillation and Self-Distil...
 
【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...
【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...
【DL輪読会】VIP: Towards Universal Visual Reward and Representation via Value-Impl...
 
【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...
【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...
【DL輪読会】Deep Transformers without Shortcuts: Modifying Self-attention for Fait...
 
【DL輪読会】マルチモーダル 基盤モデル
【DL輪読会】マルチモーダル 基盤モデル【DL輪読会】マルチモーダル 基盤モデル
【DL輪読会】マルチモーダル 基盤モデル
 
【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...
【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...
【DL輪読会】TrOCR: Transformer-based Optical Character Recognition with Pre-traine...
 
【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...
【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...
【DL輪読会】HyperDiffusion: Generating Implicit Neural Fields withWeight-Space Dif...
 
【DL輪読会】大量API・ツールの扱いに特化したLLM
【DL輪読会】大量API・ツールの扱いに特化したLLM【DL輪読会】大量API・ツールの扱いに特化したLLM
【DL輪読会】大量API・ツールの扱いに特化したLLM
 
【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision
【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision
【DL輪読会】DINOv2: Learning Robust Visual Features without Supervision
 
【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...
【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...
【DL輪読会】Poisoning Language Models During Instruction Tuning Instruction Tuning...
 

[DL輪読会]“SimGANs: Simulator-Based Generative Adversarial Networks for ECG Synthesis to Improve Deep ECG Classification (ICML2020)”

Hinweis der Redaktion

  1. https://www.kango-roo.com/ki/image_673/
  2. GANはクラスごとに作成 シミュレータの出力は、Normal Classのみ