SlideShare a Scribd company logo
1 of 14
Download to read offline
Domain adaptation
The University of Tokyo, Master 1st year
Koike Tomoya
What’s domain adaptation?
2
Train dataset Test dataset
Get robustness in different data generation distribution
Source domain Target domain
Domain adaptation
Types of domain adaptation[1]
3
Source domain
Target domain
for train
Situation
Supervised learning — Data & Label Afford the cost
Unsupervised learning Data & Label Only Data
Unlabeled data is
accessible
Domain generalization Data & Label No Data New user/subject
Learning Transferable Features with Deep Adaptation Networks[4]
4
Discrepancy loss for latter layers
where phi is feature mapping function, k is kernel function and H_k is reproducing
kernel Hilbert space
Multi-kernel maximum mean discrepancy(MK-MMD) is defined as
Domain-Adversarial Training of Neural Networks[2]
5
Gradient reversal layer prevents feature extractor from
learning domain-specific feature
Domain-Adversarial Training of Neural Networks[2]
6
In the feature space, domains are inseparable,
which means domain-invariant feature is learnt
Adversarial Discriminative Domain Adaptation(ADDA)[3]
7
Generalized architecture
for adversarial domain
adaptation
Adversarial Discriminative Domain Adaptation[3]
8
1. Train source CNN and classifier
2. Fixing source CNN weights, train target CNN and Discriminator
3. Use target CNN and pre-trained classifier when testing
Target CNN is intended to learn similar feature
representation with source CNN
Unsupervised Domain Adaptation with Residual Transfer Networks[5]
9
1. By MMD, close the distance in encoded space
2. Source classifier(fs) has residual block, which are target classifier(ft)
and residual(Δf).
3. ft is trained also with entropy minimization
This network handles the different P(Y|Z) in each domains.
Asymmetric Tri-training for Unsupervised Domain Adaptation[6]
10
1. After training encoder(F) and two source classifiers(F1 and F2), make
pseudo label to target domain
2. Train F and target classifer(Ft) with pseudo label
Maximum Classifier Discrepancy for Unsupervised Domain Adaptation[7]
11
In previous methods, class decision surface was lost by
domain adaptation due to not considering it.
Maximum Classifier Discrepancy for Unsupervised Domain Adaptation[7]
12
A. Train Generator(G) and
classifiers(F1 and F2) on
source dataset
B. Fixing G, train F1 and F2
with minus discrepancy
loss on target dataset
C. Fixing F1 and F2, train G
with discrepancy loss on
target dataset
Maximum Classifier Discrepancy for Unsupervised Domain Adaptation[7]
13
Step B Step CStep A
Using two task-specific classifiers, and training generators and them
adversarially, decision boundaries don’t lose domain info and task info
Reference
14
[1] https://www.slideshare.net/yamatookamoto5/domain-generalization-via-modelagnostic-learning-of-semantic-features
[2] https://arxiv.org/abs/1505.07818
[3] https://arxiv.org/pdf/1702.05464.pdf
[4] https://arxiv.org/pdf/1502.02791.pdf
[5] https://arxiv.org/abs/1602.04433
[6] https://arxiv.org/pdf/1702.08400.pdf
[7] https://arxiv.org/pdf/1712.02560.pdf

More Related Content

What's hot

YOLOv4: optimal speed and accuracy of object detection review
YOLOv4: optimal speed and accuracy of object detection reviewYOLOv4: optimal speed and accuracy of object detection review
YOLOv4: optimal speed and accuracy of object detection reviewLEE HOSEONG
 
Introduction to Diffusion Models
Introduction to Diffusion ModelsIntroduction to Diffusion Models
Introduction to Diffusion ModelsSangwoo Mo
 
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
 
Variational Autoencoder Tutorial
Variational Autoencoder Tutorial Variational Autoencoder Tutorial
Variational Autoencoder Tutorial Hojin Yang
 
End to-end semi-supervised object detection with soft teacher ver.1.0
End to-end semi-supervised object detection with soft teacher ver.1.0End to-end semi-supervised object detection with soft teacher ver.1.0
End to-end semi-supervised object detection with soft teacher ver.1.0taeseon ryu
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and ApplicationsEmanuele Ghelfi
 
Generative adversarial network and its applications to speech signal and natu...
Generative adversarial network and its applications to speech signal and natu...Generative adversarial network and its applications to speech signal and natu...
Generative adversarial network and its applications to speech signal and natu...宏毅 李
 
Wasserstein GAN 수학 이해하기 I
Wasserstein GAN 수학 이해하기 IWasserstein GAN 수학 이해하기 I
Wasserstein GAN 수학 이해하기 ISungbin Lim
 
Feature Engineering
Feature EngineeringFeature Engineering
Feature EngineeringHJ van Veen
 
Explicit Density Models
Explicit Density ModelsExplicit Density Models
Explicit Density ModelsSangwoo Mo
 
文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...
文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...
文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...Toru Tamaki
 
Deep Learning for Computer Vision: Generative models and adversarial training...
Deep Learning for Computer Vision: Generative models and adversarial training...Deep Learning for Computer Vision: Generative models and adversarial training...
Deep Learning for Computer Vision: Generative models and adversarial training...Universitat Politècnica de Catalunya
 
Machine Learning - Object Detection and Classification
Machine Learning - Object Detection and ClassificationMachine Learning - Object Detection and Classification
Machine Learning - Object Detection and ClassificationVikas Jain
 
【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning
【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning
【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement LearningDeep Learning JP
 
ELM: Extreme Learning Machine: Learning without iterative tuning
ELM: Extreme Learning Machine: Learning without iterative tuningELM: Extreme Learning Machine: Learning without iterative tuning
ELM: Extreme Learning Machine: Learning without iterative tuningzukun
 

What's hot (20)

YOLOv4: optimal speed and accuracy of object detection review
YOLOv4: optimal speed and accuracy of object detection reviewYOLOv4: optimal speed and accuracy of object detection review
YOLOv4: optimal speed and accuracy of object detection review
 
Introduction to Diffusion Models
Introduction to Diffusion ModelsIntroduction to Diffusion Models
Introduction to Diffusion Models
 
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
 
Wasserstein GAN
Wasserstein GANWasserstein GAN
Wasserstein GAN
 
Variational Autoencoder Tutorial
Variational Autoencoder Tutorial Variational Autoencoder Tutorial
Variational Autoencoder Tutorial
 
End to-end semi-supervised object detection with soft teacher ver.1.0
End to-end semi-supervised object detection with soft teacher ver.1.0End to-end semi-supervised object detection with soft teacher ver.1.0
End to-end semi-supervised object detection with soft teacher ver.1.0
 
GAN - Theory and Applications
GAN - Theory and ApplicationsGAN - Theory and Applications
GAN - Theory and Applications
 
Generative adversarial network and its applications to speech signal and natu...
Generative adversarial network and its applications to speech signal and natu...Generative adversarial network and its applications to speech signal and natu...
Generative adversarial network and its applications to speech signal and natu...
 
Wasserstein GAN 수학 이해하기 I
Wasserstein GAN 수학 이해하기 IWasserstein GAN 수학 이해하기 I
Wasserstein GAN 수학 이해하기 I
 
Feature Engineering
Feature EngineeringFeature Engineering
Feature Engineering
 
Explicit Density Models
Explicit Density ModelsExplicit Density Models
Explicit Density Models
 
文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...
文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...
文献紹介:Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segm...
 
Transfer Learning
Transfer LearningTransfer Learning
Transfer Learning
 
Mask R-CNN
Mask R-CNNMask R-CNN
Mask R-CNN
 
Deep Learning for Computer Vision: Generative models and adversarial training...
Deep Learning for Computer Vision: Generative models and adversarial training...Deep Learning for Computer Vision: Generative models and adversarial training...
Deep Learning for Computer Vision: Generative models and adversarial training...
 
LeNet-5
LeNet-5LeNet-5
LeNet-5
 
AlexNet
AlexNetAlexNet
AlexNet
 
Machine Learning - Object Detection and Classification
Machine Learning - Object Detection and ClassificationMachine Learning - Object Detection and Classification
Machine Learning - Object Detection and Classification
 
【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning
【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning
【DL輪読会】Contrastive Learning as Goal-Conditioned Reinforcement Learning
 
ELM: Extreme Learning Machine: Learning without iterative tuning
ELM: Extreme Learning Machine: Learning without iterative tuningELM: Extreme Learning Machine: Learning without iterative tuning
ELM: Extreme Learning Machine: Learning without iterative tuning
 

Similar to Domain adaptation

Analysis on Domain Adaptation based on different papers
Analysis on Domain Adaptation based on different papersAnalysis on Domain Adaptation based on different papers
Analysis on Domain Adaptation based on different papersharshavardhan814108
 
Chapter 10.slides
Chapter 10.slidesChapter 10.slides
Chapter 10.slideslara_ays
 
(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...
(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...
(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...Naoki Shibata
 
IEEE 802.11a Physical Layer Simulation
IEEE 802.11a Physical Layer SimulationIEEE 802.11a Physical Layer Simulation
IEEE 802.11a Physical Layer SimulationMichail Grigoropoulos
 
Nov 04 MS3
Nov 04 MS3Nov 04 MS3
Nov 04 MS3Samimvez
 
Investigation of outdoor path loss models for wireless communication in bhuj
Investigation of outdoor path loss models for wireless communication in bhujInvestigation of outdoor path loss models for wireless communication in bhuj
Investigation of outdoor path loss models for wireless communication in bhujIAEME Publication
 
Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...
Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...
Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...Universitat Politècnica de Catalunya
 
Analysis of lte_radio_parameter_in_diffe
Analysis of lte_radio_parameter_in_diffeAnalysis of lte_radio_parameter_in_diffe
Analysis of lte_radio_parameter_in_diffeMd.Akm Sahansha
 
Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1
Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1
Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1Marisa Paryasto
 
Low Profile Inverted-F-L Antenna for 5.5 GHz WiMAX Applications
Low Profile Inverted-F-L Antenna for 5.5 GHz WiMAX ApplicationsLow Profile Inverted-F-L Antenna for 5.5 GHz WiMAX Applications
Low Profile Inverted-F-L Antenna for 5.5 GHz WiMAX ApplicationsIDES Editor
 
09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas
09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas
09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyasIAESIJEECS
 
S.A.kalaiselvan- robust video data hiding at forbidden zone
S.A.kalaiselvan- robust video data hiding at forbidden zoneS.A.kalaiselvan- robust video data hiding at forbidden zone
S.A.kalaiselvan- robust video data hiding at forbidden zonekalaiselvanresearch
 
a-seminar-on-manet.pptx
a-seminar-on-manet.pptxa-seminar-on-manet.pptx
a-seminar-on-manet.pptxSujit833143
 

Similar to Domain adaptation (18)

Analysis on Domain Adaptation based on different papers
Analysis on Domain Adaptation based on different papersAnalysis on Domain Adaptation based on different papers
Analysis on Domain Adaptation based on different papers
 
Chapter 10.slides
Chapter 10.slidesChapter 10.slides
Chapter 10.slides
 
(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...
(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...
(Slides) A Method for Distributed Computaion of Semi-Optimal Multicast Tree i...
 
IEEE 802.11a Physical Layer Simulation
IEEE 802.11a Physical Layer SimulationIEEE 802.11a Physical Layer Simulation
IEEE 802.11a Physical Layer Simulation
 
Nov 04 MS3
Nov 04 MS3Nov 04 MS3
Nov 04 MS3
 
Investigation of outdoor path loss models for wireless communication in bhuj
Investigation of outdoor path loss models for wireless communication in bhujInvestigation of outdoor path loss models for wireless communication in bhuj
Investigation of outdoor path loss models for wireless communication in bhuj
 
Wn ppt (1)
Wn ppt (1)Wn ppt (1)
Wn ppt (1)
 
Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...
Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...
Transfer Learning and Domain Adaptation (DLAI D5L2 2017 UPC Deep Learning for...
 
Analysis of lte_radio_parameter_in_diffe
Analysis of lte_radio_parameter_in_diffeAnalysis of lte_radio_parameter_in_diffe
Analysis of lte_radio_parameter_in_diffe
 
Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1
Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1
Iceei2011 marisa br_fajar_intan_kuspriyanto revision 1
 
Low Profile Inverted-F-L Antenna for 5.5 GHz WiMAX Applications
Low Profile Inverted-F-L Antenna for 5.5 GHz WiMAX ApplicationsLow Profile Inverted-F-L Antenna for 5.5 GHz WiMAX Applications
Low Profile Inverted-F-L Antenna for 5.5 GHz WiMAX Applications
 
WSN PPT.ppt
WSN PPT.pptWSN PPT.ppt
WSN PPT.ppt
 
abcd
abcdabcd
abcd
 
Ch3 (1)
Ch3 (1)Ch3 (1)
Ch3 (1)
 
ch3.ppt
ch3.pptch3.ppt
ch3.ppt
 
09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas
09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas
09 23sept 8434 10235-1-ed performance (edit ari)update 17jan18tyas
 
S.A.kalaiselvan- robust video data hiding at forbidden zone
S.A.kalaiselvan- robust video data hiding at forbidden zoneS.A.kalaiselvan- robust video data hiding at forbidden zone
S.A.kalaiselvan- robust video data hiding at forbidden zone
 
a-seminar-on-manet.pptx
a-seminar-on-manet.pptxa-seminar-on-manet.pptx
a-seminar-on-manet.pptx
 

Recently uploaded

biology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGYbiology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGY1301aanya
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsSérgio Sacani
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.Cherry
 
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLGwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLkantirani197
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .Poonam Aher Patil
 
Dr. E. Muralinath_ Blood indices_clinical aspects
Dr. E. Muralinath_ Blood indices_clinical  aspectsDr. E. Muralinath_ Blood indices_clinical  aspects
Dr. E. Muralinath_ Blood indices_clinical aspectsmuralinath2
 
Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Cherry
 
FAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical ScienceFAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical ScienceAlex Henderson
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learninglevieagacer
 
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptxClimate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptxDiariAli
 
PODOCARPUS...........................pptx
PODOCARPUS...........................pptxPODOCARPUS...........................pptx
PODOCARPUS...........................pptxCherry
 
Genome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptxGenome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptxCherry
 
POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.Cherry
 
Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxMohamedFarag457087
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.takadzanijustinmaime
 
Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Cherry
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptxCherry
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxseri bangash
 

Recently uploaded (20)

biology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGYbiology HL practice questions IB BIOLOGY
biology HL practice questions IB BIOLOGY
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.LUNULARIA -features, morphology, anatomy ,reproduction etc.
LUNULARIA -features, morphology, anatomy ,reproduction etc.
 
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRLGwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
Gwalior ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Gwalior ESCORT SERVICE❤CALL GIRL
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .
 
Early Development of Mammals (Mouse and Human).pdf
Early Development of Mammals (Mouse and Human).pdfEarly Development of Mammals (Mouse and Human).pdf
Early Development of Mammals (Mouse and Human).pdf
 
Dr. E. Muralinath_ Blood indices_clinical aspects
Dr. E. Muralinath_ Blood indices_clinical  aspectsDr. E. Muralinath_ Blood indices_clinical  aspects
Dr. E. Muralinath_ Blood indices_clinical aspects
 
Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.Selaginella: features, morphology ,anatomy and reproduction.
Selaginella: features, morphology ,anatomy and reproduction.
 
FAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical ScienceFAIRSpectra - Enabling the FAIRification of Analytical Science
FAIRSpectra - Enabling the FAIRification of Analytical Science
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learning
 
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptxClimate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
Climate Change Impacts on Terrestrial and Aquatic Ecosystems.pptx
 
PODOCARPUS...........................pptx
PODOCARPUS...........................pptxPODOCARPUS...........................pptx
PODOCARPUS...........................pptx
 
Genome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptxGenome sequencing,shotgun sequencing.pptx
Genome sequencing,shotgun sequencing.pptx
 
POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.POGONATUM : morphology, anatomy, reproduction etc.
POGONATUM : morphology, anatomy, reproduction etc.
 
Digital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptxDigital Dentistry.Digital Dentistryvv.pptx
Digital Dentistry.Digital Dentistryvv.pptx
 
FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.FS P2 COMBO MSTA LAST PUSH past exam papers.
FS P2 COMBO MSTA LAST PUSH past exam papers.
 
Plasmid: types, structure and functions.
Plasmid: types, structure and functions.Plasmid: types, structure and functions.
Plasmid: types, structure and functions.
 
Clean In Place(CIP).pptx .
Clean In Place(CIP).pptx                 .Clean In Place(CIP).pptx                 .
Clean In Place(CIP).pptx .
 
Human genetics..........................pptx
Human genetics..........................pptxHuman genetics..........................pptx
Human genetics..........................pptx
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptx
 

Domain adaptation

  • 1. Domain adaptation The University of Tokyo, Master 1st year Koike Tomoya
  • 2. What’s domain adaptation? 2 Train dataset Test dataset Get robustness in different data generation distribution Source domain Target domain Domain adaptation
  • 3. Types of domain adaptation[1] 3 Source domain Target domain for train Situation Supervised learning — Data & Label Afford the cost Unsupervised learning Data & Label Only Data Unlabeled data is accessible Domain generalization Data & Label No Data New user/subject
  • 4. Learning Transferable Features with Deep Adaptation Networks[4] 4 Discrepancy loss for latter layers where phi is feature mapping function, k is kernel function and H_k is reproducing kernel Hilbert space Multi-kernel maximum mean discrepancy(MK-MMD) is defined as
  • 5. Domain-Adversarial Training of Neural Networks[2] 5 Gradient reversal layer prevents feature extractor from learning domain-specific feature
  • 6. Domain-Adversarial Training of Neural Networks[2] 6 In the feature space, domains are inseparable, which means domain-invariant feature is learnt
  • 7. Adversarial Discriminative Domain Adaptation(ADDA)[3] 7 Generalized architecture for adversarial domain adaptation
  • 8. Adversarial Discriminative Domain Adaptation[3] 8 1. Train source CNN and classifier 2. Fixing source CNN weights, train target CNN and Discriminator 3. Use target CNN and pre-trained classifier when testing Target CNN is intended to learn similar feature representation with source CNN
  • 9. Unsupervised Domain Adaptation with Residual Transfer Networks[5] 9 1. By MMD, close the distance in encoded space 2. Source classifier(fs) has residual block, which are target classifier(ft) and residual(Δf). 3. ft is trained also with entropy minimization This network handles the different P(Y|Z) in each domains.
  • 10. Asymmetric Tri-training for Unsupervised Domain Adaptation[6] 10 1. After training encoder(F) and two source classifiers(F1 and F2), make pseudo label to target domain 2. Train F and target classifer(Ft) with pseudo label
  • 11. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation[7] 11 In previous methods, class decision surface was lost by domain adaptation due to not considering it.
  • 12. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation[7] 12 A. Train Generator(G) and classifiers(F1 and F2) on source dataset B. Fixing G, train F1 and F2 with minus discrepancy loss on target dataset C. Fixing F1 and F2, train G with discrepancy loss on target dataset
  • 13. Maximum Classifier Discrepancy for Unsupervised Domain Adaptation[7] 13 Step B Step CStep A Using two task-specific classifiers, and training generators and them adversarially, decision boundaries don’t lose domain info and task info
  • 14. Reference 14 [1] https://www.slideshare.net/yamatookamoto5/domain-generalization-via-modelagnostic-learning-of-semantic-features [2] https://arxiv.org/abs/1505.07818 [3] https://arxiv.org/pdf/1702.05464.pdf [4] https://arxiv.org/pdf/1502.02791.pdf [5] https://arxiv.org/abs/1602.04433 [6] https://arxiv.org/pdf/1702.08400.pdf [7] https://arxiv.org/pdf/1712.02560.pdf