SlideShare a Scribd company logo
1 of 16
Download to read offline
Interaction Networks for
Learning about Objects,
Relations and Physics
Peter Battaglia, Razvan Pascanu,
Matthew Lai, Danilo Jimenez Rezende,
koray kavukcuoglu (Google DeepMind)
NIPS 2016 Reading Club

Presenter: Ken Kuroki (@enuroi)
1
Background & Purpose
• Some attempts to learn physical dynamics so far.

(rigid bodies, fluid dynamics, 3D trajectory etc.)
• This study aims to construct a general-purpose
learnable physics engine.

(that can learn novel physical systems)
2
Model at a Glance
3
O1
O2
O1,t O2,t r
fR
et+1
O2,t
fO
et+1
O2,t+1
Model in Detail 1
4
Rr =
0 0
1 1
0 0
Rs =
1 0
0 0
0 1
Model in Detail 2
5
NR : number of relations
NO : number of objects
bk : <oi, oj, rk>

(rearranges the objects and relations into interaction terms)
Relation

e: multiple for one object
c: aggregated by a
Implementation 1
6
O = Ds
NO
R =
NR
NO
NR
NO
Rr Rs
receiver sender
DR
NR
Ra
attributes
, ,
object1's status vector
Implementation 2
7
m(G) = Ds
Ds
DR
NR
ORr
ORs
Ra
= B
[b1, b2, ..., bk]
[e1, e2, ..., ek] = E
fR
Implementation 3
8
G, X, E
E = ERr
– T
[O; X; E] = C
–
Ds
Ds
DR
NR
O
X
E
–
fR
a
P = Ot+1
DA
fA
(Free energy)
Architecture
• MLP (bias, ReLU)
By hyperparamerter search...
• FR : four 150-length hidden layers, output length 50
• FO : one 100-length hidden layer, output length 2

(x and y velocity)
• FA : one 25-length hidden layer
9
Optimization
• Used Adam

Learning rate 0.001, and downscaled by *0.8 for 40
epochs
• L2 regularization

(penalty factor by grid search)
10
Training
Simulated 2000 scenes over 1000 time steps
• Training : 1 million sample, for 2000 epochs (mini-
batches of 100 to balance distributions)
• Validation : 200k sample
• Test data : 200k sample
11
Experiments
1. N-body
2. Bouncing balls
3. String
12
Comparison
Alternative Models:
1. Constant velocity (output=input)
2. MLP (two 300-length hidden layers)

input: flattened vector of all the input data
3. Interaction Network without E (interaction)
13
Results
14
Discussion
1. Performed better than alternatives
2. Baseline MLP couldn't effectively learn interaction
3. To understand "intuitive physics engine" in human
4. Potential to expand the model
15
Presenter's Comments
1. Can be applied to a larger system?

(time & memory-wise)
2. Probably it can be parallelized
3. Really advantageous to alternatives?
16

More Related Content

What's hot

Generative adversarial networks
Generative adversarial networksGenerative adversarial networks
Generative adversarial networksYunjey Choi
 
Recurrent Neural Networks
Recurrent Neural NetworksRecurrent Neural Networks
Recurrent Neural NetworksSeiya Tokui
 
計算スケジューリングの効果~もし,Halideがなかったら?~
計算スケジューリングの効果~もし,Halideがなかったら?~計算スケジューリングの効果~もし,Halideがなかったら?~
計算スケジューリングの効果~もし,Halideがなかったら?~Norishige Fukushima
 
1시간만에 GAN(Generative Adversarial Network) 완전 정복하기
1시간만에 GAN(Generative Adversarial Network) 완전 정복하기1시간만에 GAN(Generative Adversarial Network) 완전 정복하기
1시간만에 GAN(Generative Adversarial Network) 완전 정복하기NAVER Engineering
 
[논문리뷰] 딥러닝 적용한 EEG 연구 소개
[논문리뷰] 딥러닝 적용한 EEG 연구 소개[논문리뷰] 딥러닝 적용한 EEG 연구 소개
[논문리뷰] 딥러닝 적용한 EEG 연구 소개Donghyeon Kim
 
量子アニーリングマシンのプログラミング
量子アニーリングマシンのプログラミング量子アニーリングマシンのプログラミング
量子アニーリングマシンのプログラミングnishio
 
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...SSII
 
(DL輪読)Variational Dropout Sparsifies Deep Neural Networks
(DL輪読)Variational Dropout Sparsifies Deep Neural Networks(DL輪読)Variational Dropout Sparsifies Deep Neural Networks
(DL輪読)Variational Dropout Sparsifies Deep Neural NetworksMasahiro Suzuki
 
Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)Yoshitaka Ushiku
 
Variational Autoencoder
Variational AutoencoderVariational Autoencoder
Variational AutoencoderMark Chang
 
강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)
강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)
강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)Euijin Jeong
 
Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10) Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10) Larry Guo
 
SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向
SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向
SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向SSII
 
Graph Attention Network
Graph Attention NetworkGraph Attention Network
Graph Attention NetworkTakahiro Kubo
 
Subword tokenizers
Subword tokenizersSubword tokenizers
Subword tokenizersHa Loc Do
 
最近の自然言語処理
最近の自然言語処理最近の自然言語処理
最近の自然言語処理naoto moriyama
 
遺伝的アルゴリズム・遺伝的プログラミング
遺伝的アルゴリズム・遺伝的プログラミング遺伝的アルゴリズム・遺伝的プログラミング
遺伝的アルゴリズム・遺伝的プログラミングMatsuiRyo
 
Information-Theoretic Metric Learning
Information-Theoretic Metric LearningInformation-Theoretic Metric Learning
Information-Theoretic Metric LearningKoji Matsuda
 

What's hot (20)

Generative adversarial networks
Generative adversarial networksGenerative adversarial networks
Generative adversarial networks
 
Recurrent Neural Networks
Recurrent Neural NetworksRecurrent Neural Networks
Recurrent Neural Networks
 
計算スケジューリングの効果~もし,Halideがなかったら?~
計算スケジューリングの効果~もし,Halideがなかったら?~計算スケジューリングの効果~もし,Halideがなかったら?~
計算スケジューリングの効果~もし,Halideがなかったら?~
 
1시간만에 GAN(Generative Adversarial Network) 완전 정복하기
1시간만에 GAN(Generative Adversarial Network) 완전 정복하기1시간만에 GAN(Generative Adversarial Network) 완전 정복하기
1시간만에 GAN(Generative Adversarial Network) 완전 정복하기
 
[논문리뷰] 딥러닝 적용한 EEG 연구 소개
[논문리뷰] 딥러닝 적용한 EEG 연구 소개[논문리뷰] 딥러닝 적용한 EEG 연구 소개
[논문리뷰] 딥러닝 적용한 EEG 연구 소개
 
量子アニーリングマシンのプログラミング
量子アニーリングマシンのプログラミング量子アニーリングマシンのプログラミング
量子アニーリングマシンのプログラミング
 
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
SSII2021 [SS1] Transformer x Computer Visionの 実活用可能性と展望 〜 TransformerのCompute...
 
imgproxy is amazing
imgproxy is amazingimgproxy is amazing
imgproxy is amazing
 
(DL輪読)Variational Dropout Sparsifies Deep Neural Networks
(DL輪読)Variational Dropout Sparsifies Deep Neural Networks(DL輪読)Variational Dropout Sparsifies Deep Neural Networks
(DL輪読)Variational Dropout Sparsifies Deep Neural Networks
 
Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)Curriculum Learning (関東CV勉強会)
Curriculum Learning (関東CV勉強会)
 
Variational Autoencoder
Variational AutoencoderVariational Autoencoder
Variational Autoencoder
 
강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)
강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)
강화학습 기초_2(Deep sarsa, Deep Q-learning, DQN)
 
Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10) Deep Learning: Recurrent Neural Network (Chapter 10)
Deep Learning: Recurrent Neural Network (Chapter 10)
 
SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向
SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向
SSII2021 [OS2-02] 深層学習におけるデータ拡張の原理と最新動向
 
Graph Attention Network
Graph Attention NetworkGraph Attention Network
Graph Attention Network
 
Subword tokenizers
Subword tokenizersSubword tokenizers
Subword tokenizers
 
最近の自然言語処理
最近の自然言語処理最近の自然言語処理
最近の自然言語処理
 
遺伝的アルゴリズム・遺伝的プログラミング
遺伝的アルゴリズム・遺伝的プログラミング遺伝的アルゴリズム・遺伝的プログラミング
遺伝的アルゴリズム・遺伝的プログラミング
 
DeepLearningTutorial
DeepLearningTutorialDeepLearningTutorial
DeepLearningTutorial
 
Information-Theoretic Metric Learning
Information-Theoretic Metric LearningInformation-Theoretic Metric Learning
Information-Theoretic Metric Learning
 

Viewers also liked

Introduction of "TrailBlazer" algorithm
Introduction of "TrailBlazer" algorithmIntroduction of "TrailBlazer" algorithm
Introduction of "TrailBlazer" algorithmKatsuki Ohto
 
時系列データ3
時系列データ3時系列データ3
時系列データ3graySpace999
 
Fast and Probvably Seedings for k-Means
Fast and Probvably Seedings for k-MeansFast and Probvably Seedings for k-Means
Fast and Probvably Seedings for k-MeansKimikazu Kato
 
Dual Learning for Machine Translation (NIPS 2016)
Dual Learning for Machine Translation (NIPS 2016)Dual Learning for Machine Translation (NIPS 2016)
Dual Learning for Machine Translation (NIPS 2016)Toru Fujino
 
InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...Shuhei Yoshida
 
Safe and Efficient Off-Policy Reinforcement Learning
Safe and Efficient Off-Policy Reinforcement LearningSafe and Efficient Off-Policy Reinforcement Learning
Safe and Efficient Off-Policy Reinforcement Learningmooopan
 
Learning to learn by gradient descent by gradient descent
Learning to learn by gradient descent by gradient descentLearning to learn by gradient descent by gradient descent
Learning to learn by gradient descent by gradient descentHiroyuki Fukuda
 
Introduction of “Fairness in Learning: Classic and Contextual Bandits”
Introduction of “Fairness in Learning: Classic and Contextual Bandits”Introduction of “Fairness in Learning: Classic and Contextual Bandits”
Introduction of “Fairness in Learning: Classic and Contextual Bandits”Kazuto Fukuchi
 
Conditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN DecodersConditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN Decoderssuga93
 
Improving Variational Inference with Inverse Autoregressive Flow
Improving Variational Inference with Inverse Autoregressive FlowImproving Variational Inference with Inverse Autoregressive Flow
Improving Variational Inference with Inverse Autoregressive FlowTatsuya Shirakawa
 
[DL輪読会]Convolutional Sequence to Sequence Learning
[DL輪読会]Convolutional Sequence to Sequence Learning[DL輪読会]Convolutional Sequence to Sequence Learning
[DL輪読会]Convolutional Sequence to Sequence LearningDeep Learning JP
 
論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...
論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...
論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...Kusano Hitoshi
 
NIPS 2016 Overview and Deep Learning Topics
NIPS 2016 Overview and Deep Learning Topics  NIPS 2016 Overview and Deep Learning Topics
NIPS 2016 Overview and Deep Learning Topics Koichi Hamada
 
Differential privacy without sensitivity [NIPS2016読み会資料]
Differential privacy without sensitivity [NIPS2016読み会資料]Differential privacy without sensitivity [NIPS2016読み会資料]
Differential privacy without sensitivity [NIPS2016読み会資料]Kentaro Minami
 
Matching networks for one shot learning
Matching networks for one shot learningMatching networks for one shot learning
Matching networks for one shot learningKazuki Fujikawa
 
ICML2016読み会 概要紹介
ICML2016読み会 概要紹介ICML2016読み会 概要紹介
ICML2016読み会 概要紹介Kohei Hayashi
 
論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural Networks論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural NetworksSeiya Tokui
 

Viewers also liked (18)

Introduction of "TrailBlazer" algorithm
Introduction of "TrailBlazer" algorithmIntroduction of "TrailBlazer" algorithm
Introduction of "TrailBlazer" algorithm
 
時系列データ3
時系列データ3時系列データ3
時系列データ3
 
Fast and Probvably Seedings for k-Means
Fast and Probvably Seedings for k-MeansFast and Probvably Seedings for k-Means
Fast and Probvably Seedings for k-Means
 
Value iteration networks
Value iteration networksValue iteration networks
Value iteration networks
 
Dual Learning for Machine Translation (NIPS 2016)
Dual Learning for Machine Translation (NIPS 2016)Dual Learning for Machine Translation (NIPS 2016)
Dual Learning for Machine Translation (NIPS 2016)
 
InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...
InfoGAN: Interpretable Representation Learning by Information Maximizing Gen...
 
Safe and Efficient Off-Policy Reinforcement Learning
Safe and Efficient Off-Policy Reinforcement LearningSafe and Efficient Off-Policy Reinforcement Learning
Safe and Efficient Off-Policy Reinforcement Learning
 
Learning to learn by gradient descent by gradient descent
Learning to learn by gradient descent by gradient descentLearning to learn by gradient descent by gradient descent
Learning to learn by gradient descent by gradient descent
 
Introduction of “Fairness in Learning: Classic and Contextual Bandits”
Introduction of “Fairness in Learning: Classic and Contextual Bandits”Introduction of “Fairness in Learning: Classic and Contextual Bandits”
Introduction of “Fairness in Learning: Classic and Contextual Bandits”
 
Conditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN DecodersConditional Image Generation with PixelCNN Decoders
Conditional Image Generation with PixelCNN Decoders
 
Improving Variational Inference with Inverse Autoregressive Flow
Improving Variational Inference with Inverse Autoregressive FlowImproving Variational Inference with Inverse Autoregressive Flow
Improving Variational Inference with Inverse Autoregressive Flow
 
[DL輪読会]Convolutional Sequence to Sequence Learning
[DL輪読会]Convolutional Sequence to Sequence Learning[DL輪読会]Convolutional Sequence to Sequence Learning
[DL輪読会]Convolutional Sequence to Sequence Learning
 
論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...
論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...
論文紹介 Combining Model-Based and Model-Free Updates for Trajectory-Centric Rein...
 
NIPS 2016 Overview and Deep Learning Topics
NIPS 2016 Overview and Deep Learning Topics  NIPS 2016 Overview and Deep Learning Topics
NIPS 2016 Overview and Deep Learning Topics
 
Differential privacy without sensitivity [NIPS2016読み会資料]
Differential privacy without sensitivity [NIPS2016読み会資料]Differential privacy without sensitivity [NIPS2016読み会資料]
Differential privacy without sensitivity [NIPS2016読み会資料]
 
Matching networks for one shot learning
Matching networks for one shot learningMatching networks for one shot learning
Matching networks for one shot learning
 
ICML2016読み会 概要紹介
ICML2016読み会 概要紹介ICML2016読み会 概要紹介
ICML2016読み会 概要紹介
 
論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural Networks論文紹介 Pixel Recurrent Neural Networks
論文紹介 Pixel Recurrent Neural Networks
 

Similar to Interaction Networks for Learning about Objects, Relations and Physics

ELLA LC algorithm presentation in ICIP 2016
ELLA LC algorithm presentation in ICIP 2016ELLA LC algorithm presentation in ICIP 2016
ELLA LC algorithm presentation in ICIP 2016InVID Project
 
Morgan uw maGIV v1.3 dist
Morgan uw maGIV v1.3 distMorgan uw maGIV v1.3 dist
Morgan uw maGIV v1.3 distddm314
 
03j_nov18_n2.pptClassification of Parallel Computers.pptx
03j_nov18_n2.pptClassification of Parallel Computers.pptx03j_nov18_n2.pptClassification of Parallel Computers.pptx
03j_nov18_n2.pptClassification of Parallel Computers.pptxNeeraj Singh
 
Storm surge simulation
Storm surge simulationStorm surge simulation
Storm surge simulationFei Liu
 
Week 1 Lec 1-5 with watermarking.pdf
Week 1 Lec 1-5 with watermarking.pdfWeek 1 Lec 1-5 with watermarking.pdf
Week 1 Lec 1-5 with watermarking.pdfmeghana092
 
Week_1_Lec_1-5_with_watermarking_(1).pdf
Week_1_Lec_1-5_with_watermarking_(1).pdfWeek_1_Lec_1-5_with_watermarking_(1).pdf
Week_1_Lec_1-5_with_watermarking_(1).pdfPrabhaK22
 
High-throughput computation and machine learning methods applied to materials...
High-throughput computation and machine learning methods applied to materials...High-throughput computation and machine learning methods applied to materials...
High-throughput computation and machine learning methods applied to materials...Anubhav Jain
 
Deep learning: Cutting through the Myths and Hype
Deep learning: Cutting through the Myths and HypeDeep learning: Cutting through the Myths and Hype
Deep learning: Cutting through the Myths and HypeSiby Jose Plathottam
 
Introduction to Deep Learning
Introduction to Deep LearningIntroduction to Deep Learning
Introduction to Deep LearningOleg Mygryn
 
The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...
The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...
The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...PyData
 
Introduction to Chainer Chemistry
Introduction to Chainer ChemistryIntroduction to Chainer Chemistry
Introduction to Chainer ChemistryPreferred Networks
 
Discovering advanced materials for energy applications (with high-throughput ...
Discovering advanced materials for energy applications (with high-throughput ...Discovering advanced materials for energy applications (with high-throughput ...
Discovering advanced materials for energy applications (with high-throughput ...Anubhav Jain
 
Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...
Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...
Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...g Edwards
 
Introduction (Part I): High-throughput computation and machine learning appli...
Introduction (Part I): High-throughput computation and machine learning appli...Introduction (Part I): High-throughput computation and machine learning appli...
Introduction (Part I): High-throughput computation and machine learning appli...Anubhav Jain
 
Data Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and ApplicationsData Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and ApplicationsAnubhav Jain
 
Deep learning for molecules, introduction to chainer chemistry
Deep learning for molecules, introduction to chainer chemistryDeep learning for molecules, introduction to chainer chemistry
Deep learning for molecules, introduction to chainer chemistryKenta Oono
 

Similar to Interaction Networks for Learning about Objects, Relations and Physics (20)

ELLA LC algorithm presentation in ICIP 2016
ELLA LC algorithm presentation in ICIP 2016ELLA LC algorithm presentation in ICIP 2016
ELLA LC algorithm presentation in ICIP 2016
 
Morgan uw maGIV v1.3 dist
Morgan uw maGIV v1.3 distMorgan uw maGIV v1.3 dist
Morgan uw maGIV v1.3 dist
 
03j_nov18_n2.pptClassification of Parallel Computers.pptx
03j_nov18_n2.pptClassification of Parallel Computers.pptx03j_nov18_n2.pptClassification of Parallel Computers.pptx
03j_nov18_n2.pptClassification of Parallel Computers.pptx
 
Storm surge simulation
Storm surge simulationStorm surge simulation
Storm surge simulation
 
Week 1 Lec 1-5 with watermarking.pdf
Week 1 Lec 1-5 with watermarking.pdfWeek 1 Lec 1-5 with watermarking.pdf
Week 1 Lec 1-5 with watermarking.pdf
 
Week_1_Lec_1-5_with_watermarking_(1).pdf
Week_1_Lec_1-5_with_watermarking_(1).pdfWeek_1_Lec_1-5_with_watermarking_(1).pdf
Week_1_Lec_1-5_with_watermarking_(1).pdf
 
High-throughput computation and machine learning methods applied to materials...
High-throughput computation and machine learning methods applied to materials...High-throughput computation and machine learning methods applied to materials...
High-throughput computation and machine learning methods applied to materials...
 
Deep learning: Cutting through the Myths and Hype
Deep learning: Cutting through the Myths and HypeDeep learning: Cutting through the Myths and Hype
Deep learning: Cutting through the Myths and Hype
 
Licentiate Defense Slide
Licentiate Defense SlideLicentiate Defense Slide
Licentiate Defense Slide
 
Introduction to Deep Learning
Introduction to Deep LearningIntroduction to Deep Learning
Introduction to Deep Learning
 
The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...
The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...
The Face of Nanomaterials: Insightful Classification Using Deep Learning - An...
 
Introduction to Chainer Chemistry
Introduction to Chainer ChemistryIntroduction to Chainer Chemistry
Introduction to Chainer Chemistry
 
5th sem
5th sem5th sem
5th sem
 
5th sem
5th sem5th sem
5th sem
 
Discovering advanced materials for energy applications (with high-throughput ...
Discovering advanced materials for energy applications (with high-throughput ...Discovering advanced materials for energy applications (with high-throughput ...
Discovering advanced materials for energy applications (with high-throughput ...
 
Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...
Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...
Operational Transformation in Real-Time Group Editors: Issues, Algorithms, an...
 
Introduction (Part I): High-throughput computation and machine learning appli...
Introduction (Part I): High-throughput computation and machine learning appli...Introduction (Part I): High-throughput computation and machine learning appli...
Introduction (Part I): High-throughput computation and machine learning appli...
 
Data Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and ApplicationsData Mining to Discovery for Inorganic Solids: Software Tools and Applications
Data Mining to Discovery for Inorganic Solids: Software Tools and Applications
 
Deep learning for molecules, introduction to chainer chemistry
Deep learning for molecules, introduction to chainer chemistryDeep learning for molecules, introduction to chainer chemistry
Deep learning for molecules, introduction to chainer chemistry
 
Deep learning (2)
Deep learning (2)Deep learning (2)
Deep learning (2)
 

Recently uploaded

Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Victor Rentea
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamUiPathCommunity
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingEdi Saputra
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Bhuvaneswari Subramani
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...apidays
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDropbox
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWERMadyBayot
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Victor Rentea
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024The Digital Insurer
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...Zilliz
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontologyjohnbeverley2021
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century educationjfdjdjcjdnsjd
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Zilliz
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobeapidays
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAndrey Devyatkin
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Jeffrey Haguewood
 

Recently uploaded (20)

Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​Elevate Developer Efficiency & build GenAI Application with Amazon Q​
Elevate Developer Efficiency & build GenAI Application with Amazon Q​
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
Apidays New York 2024 - APIs in 2030: The Risk of Technological Sleepwalk by ...
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
Modular Monolith - a Practical Alternative to Microservices @ Devoxx UK 2024
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Six Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal OntologySix Myths about Ontologies: The Basics of Formal Ontology
Six Myths about Ontologies: The Basics of Formal Ontology
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)Introduction to Multilingual Retrieval Augmented Generation (RAG)
Introduction to Multilingual Retrieval Augmented Generation (RAG)
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 

Interaction Networks for Learning about Objects, Relations and Physics

  • 1. Interaction Networks for Learning about Objects, Relations and Physics Peter Battaglia, Razvan Pascanu, Matthew Lai, Danilo Jimenez Rezende, koray kavukcuoglu (Google DeepMind) NIPS 2016 Reading Club
 Presenter: Ken Kuroki (@enuroi) 1
  • 2. Background & Purpose • Some attempts to learn physical dynamics so far.
 (rigid bodies, fluid dynamics, 3D trajectory etc.) • This study aims to construct a general-purpose learnable physics engine.
 (that can learn novel physical systems) 2
  • 3. Model at a Glance 3 O1 O2 O1,t O2,t r fR et+1 O2,t fO et+1 O2,t+1
  • 4. Model in Detail 1 4 Rr = 0 0 1 1 0 0 Rs = 1 0 0 0 0 1
  • 5. Model in Detail 2 5 NR : number of relations NO : number of objects bk : <oi, oj, rk>
 (rearranges the objects and relations into interaction terms) Relation
 e: multiple for one object c: aggregated by a
  • 6. Implementation 1 6 O = Ds NO R = NR NO NR NO Rr Rs receiver sender DR NR Ra attributes , , object1's status vector
  • 7. Implementation 2 7 m(G) = Ds Ds DR NR ORr ORs Ra = B [b1, b2, ..., bk] [e1, e2, ..., ek] = E fR
  • 8. Implementation 3 8 G, X, E E = ERr – T [O; X; E] = C – Ds Ds DR NR O X E – fR a P = Ot+1 DA fA (Free energy)
  • 9. Architecture • MLP (bias, ReLU) By hyperparamerter search... • FR : four 150-length hidden layers, output length 50 • FO : one 100-length hidden layer, output length 2
 (x and y velocity) • FA : one 25-length hidden layer 9
  • 10. Optimization • Used Adam
 Learning rate 0.001, and downscaled by *0.8 for 40 epochs • L2 regularization
 (penalty factor by grid search) 10
  • 11. Training Simulated 2000 scenes over 1000 time steps • Training : 1 million sample, for 2000 epochs (mini- batches of 100 to balance distributions) • Validation : 200k sample • Test data : 200k sample 11
  • 12. Experiments 1. N-body 2. Bouncing balls 3. String 12
  • 13. Comparison Alternative Models: 1. Constant velocity (output=input) 2. MLP (two 300-length hidden layers)
 input: flattened vector of all the input data 3. Interaction Network without E (interaction) 13
  • 15. Discussion 1. Performed better than alternatives 2. Baseline MLP couldn't effectively learn interaction 3. To understand "intuitive physics engine" in human 4. Potential to expand the model 15
  • 16. Presenter's Comments 1. Can be applied to a larger system?
 (time & memory-wise) 2. Probably it can be parallelized 3. Really advantageous to alternatives? 16