Arindam-1991 / Domain-generalization

All about domain generalization

Domain generalization

---

Table of contents

• Research papers

  • Pathfinder
  • arXiv
  • Computer vision venues
  • Machine learning venues

• DG variants

• Datasets

• References

• Contact

• License

---

Research papers

Pathfinder

• Generalizing from several related classification tasks to a new unlabeled sample
  Blanchard, Gilles, Gyemin Lee, and Clayton Scott.
  Advances in neural information processing systems. (NIPS) 2011.

arXiv

• The Risks of Invariant Risk Minimization
  Rosenfeld, Elan, Pradeep Ravikumar, and Andrej Risteski.
  arXiv preprint arXiv:2010.05761 (2020).

• (NILE) A causal framework for distribution generalization
  Christiansen, Rune, Niklas Pfister, Martin Emil Jakobsen, Nicola Gnecco, and Jonas Peters.
  arXiv preprint arXiv:2006.07433 (2020).

• (HIR) Respecting Domain Relations: Hypothesis Invariance for Domain Generalization
  Wang, Ziqi, Marco Loog, and Jan van Gemert.
  arXiv preprint arXiv:2010.07591 (2020).

• (REx) Out-of-distribution generalization via risk extrapolation
  Krueger, David, Ethan Caballero, Joern-Henrik Jacobsen, Amy Zhang, Jonathan Binas, Remi Le Priol, and Aaron Courville.
  arXiv preprint arXiv:2003.00688 (2020).

• (RVP) Risk Variance Penalization: From Distributional Robustness to Causality
  Xie, Chuanlong, Fei Chen, Yue Liu, and Zhenguo Li.
  arXiv preprint arXiv:2006.07544 (2020).

• Generalization and Invariances in the Presence of Unobserved Confounding
  Bellot, Alexis and van der Schaar, Mihaela.
  arXiv preprint arXiv:2007.10653 (2020).

• (DomainBed) In Search of Lost Domain Generalization
  Gulrajani, Ishaan, and David Lopez-Paz.
  arXiv preprint arXiv:2007.01434 (2020).
  [code] (coming soon)

• (FAR) Feature Alignment and Restoration for Domain Generalization and Adaptation
  Jin, Xin, Cuiling Lan, Wenjun Zeng, and Zhibo Chen.
  arXiv preprint arXiv:2006.12009 (2020).

• Frustratingly Simple Domain Generalization via Image Stylization
  Somavarapu, Nathan, Chih-Yao Ma, and Zsolt Kira.
  arXiv preprint arXiv:2006.11207 (2020).
  [code]

• (RVR) Representation via Representations: Domain Generalization via Adversarially Learned Invariant Representations
  Deng, Zhun, Frances Ding, Cynthia Dwork, Rachel Hong, Giovanni Parmigiani, Prasad Patil, and Pragya Sur.
  arXiv preprint arXiv:2006.11478 (2020).

• (MatchDG) Domain Generalization using Causal Matching
  Mahajan, Divyat, Shruti Tople, and Amit Sharma.
  arXiv preprint arXiv:2006.07500 (2020).
  [code]

• Adversarial target-invariant representation learning for domain generalization
  Isabela Albuquerque, João Monteiro, Mohammad Darvishi, Tiago H. Falk, Ioannis Mitliagkas
  arXiv preprint arXiv:1911.00804 (2019).
  [code]

• DIVA: Domain Invariant Variational Autoencoders
  Maximilian Ilse, Jakub M. Tomczak, Christos Louizos, Max Welling
  arXiv preprint arXiv:1905.10427 (2019).
  [code]

• Invariant Risk Minimization
  Arjovsky, Martin and Bottou, Leon and Gulrajani, Ishaan and Lopez-Paz, David.
  arXiv preprint arXiv:1907.02893 (2019).
  [code]

• A Generalization Error Bound for Multi-class Domain Generalization
  Deshmukh, Aniket Anand, Yunwen Lei, Srinagesh Sharma, Urun Dogan, James W. Cutler, and Clayton Scott.
  arXiv preprint arXiv:1905.10392 (2019).
  [code]

• Domain generalization by marginal transfer learning
  Blanchard, Gilles, Aniket Anand Deshmukh, Urun Dogan, Gyemin Lee, and Clayton Scott.
  arXiv preprint arXiv:1711.07910 (2017).
  [code]

Computer vision venues

Autoencoder-based methods

• (MMD-AAE) Domain generalization with adversarial feature learning
  Li, Haoliang, Sinno Jialin Pan, Shiqi Wang, and Alex C. Kot.
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018.

• (MTAE) Domain generalization for object recognition with multi-task autoencoders
  Ghifary, Muhammad, W. Bastiaan Kleijn, Mengjie Zhang, and David Balduzzi.
  Proceedings of the IEEE International Conference on Computer Vision (ICCV) 2015.
  [code]

Deep neural network-based methods

• (DMG) Learning to Balance Specificity and Invariance for In and Out of Domain Generalization
  Chattopadhyay, Prithvijit, Yogesh Balaji, and Judy Hoffman.
  Proceedings of the European Conference on Computer Vision (ECCV) 2020.
  [code]

• (DSON) Learning to Optimize Domain Specific Normalization for Domain Generalization
  Seonguk Seo, Yumin Suh, Dongwan Kim, Geeho Kim, Jongwoo Han and ohyung Han.
  Proceedings of the European Conference on Computer Vision (ECCV) 2020.

• (EISNet) Learning from Extrinsic and Intrinsic Supervisions for Domain Generalization
  Wang, Shujun, Lequan Yu, Caizi Li, Chi-Wing Fu, and Pheng-Ann Heng.
  Proceedings of the European Conference on Computer Vision (ECCV) 2020.
  [code]

• (MetaVIB) Learning to Learn with Variational Information Bottleneck for Domain Generalization
  Du, Yingjun, Jun Xu, Huan Xiong, Qiang Qiu, Xiantong Zhen, Cees GM Snoek, and Ling Shao.
  Proceedings of the European Conference on Computer Vision (ECCV) 2020.

• (RSC) Self-Challenging Improves Cross-Domain Generalization
  Huang, Zeyi, Haohan Wang, Eric P. Xing, and Dong Huang.
  Proceedings of the European Conference on Computer Vision (ECCV) 2020.

• (L2A-OT) Learning to Generate Novel Domains for Domain Generalization
  Zhou, Kaiyang, Yongxin Yang, Timothy Hospedales, and Tao Xiang.
  Proceedings of the European Conference on Computer Vision (ECCV) 2020.

• (SSDG) Single-Side Domain Generalization for Face Anti-Spoofing
  Jia, Yunpei, Jie Zhang, Shiguang Shan, and Xilin Chen.
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2020.
  [code]

• (Epi-FCR) Episodic Training for Domain Generalization
  Li, Da, Jianshu Zhang, Yongxin Yang, Cong Liu, Yi-Zhe Song, and Timothy M. Hospedales.
  Proceedings of the IEEE International Conference on Computer Vision (ICCV) 2019.
  [code]

• (JiGen) Domain Generalization by Solving Jigsaw Puzzles
  Carlucci, Fabio Maria, Antonio D'Innocente, Silvia Bucci, Barbara Caputo, and Tatiana Tommasi.
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2019.
  [code]

• (CIDDG) Deep Domain Generalization via Conditional Invariant Adversarial Networks
  Li, Ya, Xinmei Tian, Mingming Gong, Yajing Liu, Tongliang Liu, Kun Zhang, and Dacheng Tao.
  Proceedings of the European Conference on Computer Vision (ECCV) 2018.

• Deep Domain Generalization With Structured Low-Rank Constraint
  Ding, Zhengming, and Yun Fu.
  IEEE Transactions on Image Processing (TIP) 27.1 (2017): 304-313.

• (CCSA) Unified deep supervised domain adaptation and generalization
  Motiian, Saeid, Marco Piccirilli, Donald A. Adjeroh, and Gianfranco Doretto.
  Proceedings of the IEEE International Conference on Computer Vision (ICCV) 2017.
  [code]

• Deeper, broader and artier domain generalization
  Li, Da, Yongxin Yang, Yi-Zhe Song, and Timothy M. Hospedales.
  Proceedings of the IEEE International Conference on Computer Vision (ICCV) 2017.
  [code]

Metric learning-based methods

• (UML) Unbiased metric learning: On the utilization of multiple datasets and web images for softening bias
  Fang, Chen, Ye Xu, and Daniel N. Rockmore.
  Proceedings of the IEEE International Conference on Computer Vision (ICCV) 2013.

Support vector machine (SVM)-based methods

• (MVDG) Multi-view domain generalization for visual recognition
  Niu, Li, Wen Li, and Dong Xu.
  Proceedings of the IEEE International Conference on Computer Vision (ICCV) 2015.

• (LRE-SVM) Exploiting low-rank structure from latent domains for domain generalization
  Xu, Zheng, Wen Li, Li Niu, and Dong Xu.
  European Conference on Computer Vision (ECCV) 2014.
  [code]

• (Undo-Bias) Undoing the damage of dataset bias
  Khosla, Aditya, Tinghui Zhou, Tomasz Malisiewicz, Alexei A. Efros, and Antonio Torralba.
  European Conference on Computer Vision (ECCV) 2012.
  [code]

Machine learning venues

Neural network-based methods

• (LDDG) Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularization
  Li, Haoliang, YuFei Wang, Renjie Wan, Shiqi Wang, Tie-Qiang Li, and Alex C. Kot.
  Neural Information Processing Systems (NeurIPS) 2020.

• (CSD) Efficient Domain Generalization via Common-Specific Low-Rank Decomposition
  Vihari Piratla, Praneeth Netrapalli, Sunita Sarawagi
  International Conference on Machine Learning (ICML) 2020.
  [code]

• (GCFN) Generalized Convolutional Forest Networks for Domain Generalization and Visual Recognition.
  Ryu, Jongbin, Gitaek Kwon, Ming-Hsuan Yang, and Jongwoo Lim.
  International Conference on Learning Representations (ICLR) 2020.

• (MASF) Domain Generalization via Model-Agnostic Learning of Semantic Features.
  Qi Dou, Daniel C. Castro, Konstantinos Kamnitsas, and Ben Glocker.
  Advances in Neural Information Processing Systems (NeurIPS) 2019.
  [code]

• (CAADA) Correlation-aware Adversarial Domain Adaptation and Generalization
  Rahman, Mohammad Mahfujur, Clinton Fookes, Mahsa Baktashmotlagh, and Sridha Sridharan.
  Pattern Recognition (2019): 107124.

• (CROSSGRAD) Generalizing Across Domains via Cross-Gradient Training
  Shankar, Shiv, Vihari Piratla, Soumen Chakrabarti, Siddhartha Chaudhuri, Preethi Jyothi, and Sunita Sarawagi.
  International Conference on Learning Representations (ICLR) 2018.

• (MetaReg) MetaReg: Towards Domain Generalization using Meta-Regularization
  Balaji, Yogesh, Swami Sankaranarayanan, and Rama Chellappa.
  Advances in Neural Information Processing Systems (NeurIPS) 2018.

• (MLDG) Learning to generalize: Meta-learning for domain generalization
  Li, Da, Yongxin Yang, Yi-Zhe Song, and Timothy M. Hospedales.
  AAAI Conference on Artificial Intelligence (AAAI) 2018.
  [code]

Kernel-based methods

• (MDA) Domain Generalization via Multidomain Discriminant Analysis
  Hu, Shoubo, Kun Zhang, Zhitang Chen, Laiwan Chan.
  Conference on Uncertainty in Artificial Intelligence (UAI) 2019.
  [code]

• (CIDG) Domain Generalization via Conditional Invariant Representation
  Li, Ya, Mingming Gong, Xinmei Tian, Tongliang Liu, and Dacheng Tao.
  AAAI Conference on Artificial Intelligence (AAAI) 2018.
  [code]

• (SCA) Scatter component analysis: A unified framework for domain adaptation and domain generalization
  Ghifary, Muhammad, David Balduzzi, W. Bastiaan Kleijn, and Mengjie Zhang.
  IEEE Transactions on Pattern Analysis & Machine Intelligence (TPAMI) 39.7 (2016): 1414-1430.
  [code(unofficial)]

• (DICA) Domain generalization via invariant feature representation
  Muandet, Krikamol, David Balduzzi, and Bernhard Schölkopf.
  International Conference on Machine Learning (ICML) 2013.
  [code]

---

DG variants

• Exchanging Lessons Between Algorithmic Fairness and Domain Generalization
  Creager, Elliot, Jörn-Henrik Jacobsen, and Richard Zemel.
  arXiv preprint arXiv:2010.07249 2020.

• Learning to Learn Single Domain Generalization
  Fengchun Qiao, Long Zhao, Xi Peng.
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2020.

• Domain Generalization Using a Mixture of Multiple Latent Domains
  Toshihiko Matsuura, Tatsuya Harada.
  AAAI Conference on Artificial Intelligence (AAAI) 2020.
  [code]

• (APN) Adversarial Pyramid Network for Video Domain Generalization
  Zhiyu Yao, Yunbo Wang, Xingqiang Du, Mingsheng Long, Jianmin Wang
  arXiv preprint arXiv:1912.03716 (2019).

• (FC) Feature-Critic Networks for Heterogeneous Domain Generalization
  Li, Yiying, Yongxin Yang, Wei Zhou, and Timothy M. Hospedales
  International Conference on Machine Learning (ICML) 2019.
  [code]

• Learning Robust Representations by Projecting Superficial Statistics Out
  Wang, Haohan, Zexue He, Zachary C. Lipton, and Eric P. Xing.
  International Conference on Learning Representations (ICLR) 2019.

---

Datasets

Dataset	#Sample	#Feature	#Class	Subdomain	Reference
Office+Caltech	2533	SURF: 800, DeCAF: 4096	10	A, W, D, C	[1]
VOC2007	3376	DeCAF: 4096	5	V	[2]
LabelMe	2656	DeCAF: 4096	5	L	[3]
Caltech101	1415	DeCAF: 4096	5	C	[4]
SUN09	3282	DeCAF: 4096	5	S	[5]
PACS	9991	ResNet: 512, AlexNet: 4096	7	Photo, Art Painting, Cartoon, Sketch	[6]

Office+Caltech

Introduction

This dataset is constructed by collecting common classes in two datasets: Office-31 (which contains A, W and D) and Caltech-256 (which is C).
Four domains: A(Amazon, 958 instances), W(Webcam, 295 instances), D(DSLR, 157 instances), and C(Caltech, 1123 instances).
Ten common classes: back pack, bike, calculator, headphones, keyboard, laptop_computer, monitor, mouse, mug, and projector.

Download

Download Office+Caltech original images [Google Drive]
Download Office+Caltech SURF dataset [Google Drive]
Download Office+Caltech DeCAF dataset [Google Drive]

VLCS

Introduction

Four domains: V(VOC2007), L(LabelMe), C(Caltech), and S(SUN09).
Five common classes: bird, car, chair, dog, and person.

Download

Download the VLCS DeCAF dataset [Google Drive]

ImageNet-C

Introduction

Fifteen Corruptions spanning noise, blur, weather, and digital corruptions. 1000 common classes, the ImageNet-1K classes. The paper is here.

Download

Download links are available at https://github.com/hendrycks/robustness/

ImageNet-R

Introduction

ImageNet-R(endition) contains art, cartoons, deviantart, graffiti, embroidery, graphics, origami, paintings, patterns, plastic objects, plush objects, sculptures, sketches, tattoos, toys, and video game renditions of ImageNet classes.

ImageNet-R has renditions of 200 ImageNet classes resulting in 30,000 images. The paper is here.

Download

Download links are available at https://github.com/hendrycks/imagenet-r

PACS

Introduction

Four domains: photo, art painting, cartoon, and sketch.
Seven common classes: dog, elephant, horse, giraffe, guitar, house, and person.

Download

Download the PACS dataset [Google Drive]

---

References

1. Gong, Boqing, Yuan Shi, Fei Sha, and Kristen Grauman. "Geodesic flow kernel for unsupervised domain adaptation." In Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on, pp. 2066-2073. IEEE, 2012.

2. Everingham, Mark, Luc Van Gool, Christopher KI Williams, John Winn, and Andrew Zisserman. "The pascal visual object classes (voc) challenge." International journal of computer vision 88, no. 2 (2010): 303-338.

3. Russell, Bryan C., Antonio Torralba, Kevin P. Murphy, and William T. Freeman. "LabelMe: a database and web-based tool for image annotation." International journal of computer vision 77, no. 1-3 (2008): 157-173.

4. Griffin, Gregory, Alex Holub, and Pietro Perona. "Caltech-256 object category dataset." (2007).

5. Choi, Myung Jin, Joseph J. Lim, Antonio Torralba, and Alan S. Willsky. "Exploiting hierarchical context on a large database of object categories." (2010).

6. Da Li, Yongxin Yang, Yi-Zhe Song, and Timothy M Hospedales. "Deeper, broader and artier domaingeneralization." InProceedings of the IEEE international conference on computer vision, pages 5542–5550,2017.10. (2017).

---

Contact

• Shoubo Hu - shoubo.sub [at] gmail.com

See also the list of contributors who participated in this project.

---

License

This project is licensed under the MIT License - see the LICENSE file for details.