Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence

Diane J. Cook; Garrett Wilson

arxiv: 1907.07802 · v1 · pith:SOQPZRXLnew · submitted 2019-07-17 · 💻 cs.LG · stat.ML

Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence

Garrett Wilson , Diane J. Cook This is my paper

Pith reviewed 2026-05-24 20:11 UTC · model grok-4.3

classification 💻 cs.LG stat.ML

keywords domain adaptationpseudo labelingdiscriminatorconfidence estimationadversarial training

0 comments

The pith

Domain adaptation discriminators can also supply confidence scores for pseudo-labeling target samples.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Domain adaptation trains a discriminator to produce feature representations that ignore whether data comes from the labeled source or unlabeled target. Pseudo-labeling then assigns guessed labels to target samples and keeps only the high-confidence ones. This paper shows the discriminator's existing probability output can serve as that confidence score. The result is a single component performing both the invariance task and the selection task without adding new modules.

Core claim

The discriminator already trained to classify source versus target domains produces output probabilities that can be reused directly as weights or filters when assigning and retaining pseudo-labels on target data.

What carries the argument

The multi-purposed discriminator whose domain-classification probability is reused as pseudo-labeling confidence.

If this is right

Training pipelines need one fewer specialized head or loss term.
Pseudo-label selection becomes coupled to the same adversarial signal used for domain invariance.
Standard domain-adaptation benchmarks can be run with the combined objective.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same probability might replace separate entropy or softmax-max confidence estimators in other semi-supervised settings.
If the correlation holds, it could simplify architectures that currently maintain separate confidence networks.

Load-bearing premise

The discriminator probability for a target sample being from the source domain is a reliable indicator that the sample's pseudo-label is correct.

What would settle it

An experiment showing that discriminator probabilities have near-zero correlation with actual correctness of pseudo-labels on target data would falsify the claim.

Figures

Figures reproduced from arXiv: 1907.07802 by Diane J. Cook, Garrett Wilson.

**Figure 2.** Figure 2: After each DANN training step (Figure 1a), target [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

read the original abstract

Often domain adaptation is performed using a discriminator (domain classifier) to learn domain-invariant feature representations so that a classifier trained on labeled source data will generalize well to unlabeled target data. A line of research stemming from semi-supervised learning uses pseudo labeling to directly generate "pseudo labels" for the unlabeled target data and trains a classifier on the now-labeled target data, where the samples are selected or weighted based on some measure of confidence. In this paper, we propose multi-purposing the discriminator to not only aid in producing domain-invariant representations but also to provide pseudo labeling confidence.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Reusing the domain discriminator for pseudo-label confidence is a simple idea but the required correlation isn't justified or tested.

read the letter

The one thing to know is that the paper proposes using the domain discriminator's output as a confidence measure for pseudo-labeling in domain adaptation, on top of its usual role in learning invariant features. This is presented as a new combination of existing techniques. The paper does well by highlighting a potential way to avoid training an additional model for confidence estimation, which could simplify the pipeline a bit. However, the soft spots are significant given what's shown. The core assumption is that the probability from the domain classifier will indicate whether a pseudo-label is likely correct. But the discriminator only sees domain information during training, so its scores track domain similarity, not prediction accuracy on the target classes. These can easily diverge—a target sample might look very source-like yet be misclassified, or the opposite. The abstract offers no derivation, no toy example, and no experiments to support the link, so the idea stays speculative. The citation pattern isn't an issue since it's building on standard lines, but without results it's impossible to judge if this multi-purposing actually improves anything over standard pseudo-labeling with entropy or other confidences. This is for people actively working on semi-supervised domain adaptation who might want to try small modifications. A casual reader in the broader field won't get much out of it. It doesn't show clear thinking on the mechanics because the key correlation is left unexamined. I wouldn't bring it to a reading group or cite it. It doesn't deserve peer review yet; the authors would need to add validation or a theoretical argument before it merits referee time.

Referee Report

2 major / 0 minor

Summary. The manuscript proposes multi-purposing the domain discriminator in unsupervised domain adaptation: it is used both to produce domain-invariant feature representations via adversarial training and to supply a confidence score (its domain-classification probability) for selecting or weighting pseudo-labels generated by a source-trained classifier on unlabeled target data.

Significance. If the discriminator output proves to be a reliable proxy for pseudo-label correctness, the method would eliminate the need for a separate confidence estimator in pseudo-labeling pipelines and integrate directly with existing adversarial DA frameworks, offering a lightweight way to combine the two lines of work. The idea is conceptually economical but currently lacks any supporting derivation or evidence.

major comments (2)

[Abstract] Abstract: the central claim requires that p(domain|features) correlates with correctness of the source classifier's pseudo-label on a target sample. No inductive bias, derivation, or even a toy argument is supplied to establish this correlation; the discriminator is trained exclusively to distinguish source from target, a task orthogonal to class-conditional prediction error.
[Abstract] Abstract: because the manuscript consists only of the high-level proposal with no equations, algorithm, experiments, or error analysis, it is impossible to verify whether the proposed reuse of the discriminator actually supports the stated claim or merely restates an untested assumption.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their comments on our manuscript. We address each major comment below.

read point-by-point responses

Referee: [Abstract] Abstract: the central claim requires that p(domain|features) correlates with correctness of the source classifier's pseudo-label on a target sample. No inductive bias, derivation, or even a toy argument is supplied to establish this correlation; the discriminator is trained exclusively to distinguish source from target, a task orthogonal to class-conditional prediction error.

Authors: We agree that the manuscript does not provide a derivation or toy argument for why the domain discriminator's output would correlate with pseudo-label correctness. The proposal is based on the intuition that the discriminator learns features that capture domain-specific characteristics which may overlap with classification difficulty, but this is indeed an unproven assumption in the current text. We will add a section discussing this potential correlation and its limitations in the revised manuscript. revision: yes
Referee: [Abstract] Abstract: because the manuscript consists only of the high-level proposal with no equations, algorithm, experiments, or error analysis, it is impossible to verify whether the proposed reuse of the discriminator actually supports the stated claim or merely restates an untested assumption.

Authors: The manuscript is presented as a conceptual idea for multi-purposing the discriminator. We acknowledge the absence of detailed equations, algorithms, and experiments, which limits the ability to empirically verify the claim. In response, we will expand the manuscript to include a formal description of the method, pseudocode, and initial experimental results on benchmark datasets to substantiate the proposal. revision: yes

Circularity Check

0 steps flagged

No circularity: proposal is a methodological suggestion without reduction to fitted inputs or self-definitions

full rationale

The paper proposes repurposing a domain discriminator (trained to distinguish source vs. target) to also supply pseudo-label confidence scores for target samples. No equations, fitting procedures, or derivation chains are visible in the abstract or described text that would make the claimed confidence measure equivalent to its inputs by construction. The central step is an empirical hypothesis that domain-classification probability correlates with pseudo-label correctness; this is presented as a novel multi-use rather than derived from prior fitted parameters or self-citations. The work remains self-contained as an architectural suggestion without load-bearing self-referential reductions.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no explicit free parameters, axioms, or invented entities; the proposal implicitly rests on standard domain-adaptation assumptions that the discriminator can be trained to separate domains and that its output correlates with label reliability.

pith-pipeline@v0.9.0 · 5613 in / 972 out tokens · 15990 ms · 2026-05-24T20:11:38.113124+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

61 extracted references · 61 canonical work pages · 5 internal anchors

[1]

Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette, and Mario Marchand. 2014. Domain-adversarial neural networks. arXiv preprint arXiv:1412.4446 (2014)

work page internal anchor Pith review Pith/arXiv arXiv 2014
[2]

Massih-Reza Amini and Patrick Gallinari. 2002. Semi-supervised logistic regres- sion. In ECAI. 390–394. Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence AdvML’19: Workshop on Adversarial Learning Methods for Machine Learning and Data Mining at KDD, August 5th, 2019, Anchorage, Alaska, USA

work page 2002
[3]

Oscar Beijbom. 2012. Domain adaptations for computer vision applications.arXiv preprint arXiv:1211.4860 (2012)

work page internal anchor Pith review Pith/arXiv arXiv 2012
[4]

Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman Vaughan. 2010. A theory of learning from different domains. Machine Learning 79, 1 (01 May 2010), 151–175. https://doi.org/10.1007/s10994- 009-5152-4

work page doi:10.1007/s10994- 2010
[5]

Konstantinos Bousmalis, George Trigeorgis, Nathan Silberman, Dilip Krishnan, and Dumitru Erhan. 2016. Domain Separation Networks. In Advances in Neural Information Processing Systems 29, D. D. Lee, M. Sugiyama, U. V. Luxburg, I. Guyon, and R. Garnett (Eds.). Curran Associates, Inc., 343–351. http://papers.nips.cc/ paper/6254-domain-separation-networks.pdf

work page 2016
[6]

Fabio Maria Carlucci, Lorenzo Porzi, Barbara Caputo, Elisa Ricci, and Samuel Rota Bulò. 2017. Autodial: Automatic domain alignment layers. In 2017 IEEE Interna- tional Conference on Computer Vision (ICCV) . IEEE, 5077–5085

work page 2017
[7]

Olivier Chapelle and Alexander Zien. 2005. Semi-supervised classification by low density separation.. In AISTATS, Vol. 2005. Citeseer, 57–64

work page 2005
[8]

Minmin Chen, Kilian Q Weinberger, and John Blitzer. 2011. Co-Training for Domain Adaptation. In Advances in Neural Information Processing Systems 24 , J. Shawe-Taylor, R. S. Zemel, P. L. Bartlett, F. Pereira, and K. Q. Weinberger (Eds.). Curran Associates, Inc., 2456–2464. http://papers.nips.cc/paper/4433-co- training-for-domain-adaptation.pdf

work page 2011
[9]

Nicolas Courty, Rémi Flamary, Amaury Habrard, and Alain Rakotomamonjy. 2017. Joint distribution optimal transportation for domain adaptation. In Advances in Neural Information Processing Systems 30 , I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.). Curran Associates, Inc., 3730–3739. http://papers.nips.cc/...

work page 2017
[10]

Bharath Bhushan Damodaran, Benjamin Kellenberger, Rémi Flamary, Devis Tuia, and Nicolas Courty. 2018. DeepJDOT: Deep Joint Distribution Optimal Transport for Unsupervised Domain Adaptation. In Computer Vision – ECCV 2018 , Vittorio Ferrari, Martial Hebert, Cristian Sminchisescu, and Yair Weiss (Eds.). Springer International Publishing, Cham, 467–483

work page 2018
[11]

Debasmit Das and C. S. George Lee. 2018. Graph Matching and Pseudo-Label Guided Deep Unsupervised Domain Adaptation. InArtificial Neural Networks and Machine Learning – ICANN 2018 , Věra Kůrková, Yannis Manolopoulos, Barbara Hammer, Lazaros Iliadis, and Ilias Maglogiannis (Eds.). Springer International Publishing, Cham, 342–352

work page 2018
[12]

Hal Daumé III. 2012. A course in machine learning. Publisher, ciml. info 5 (2012), 69

work page 2012
[13]

Geoff French, Michal Mackiewicz, and Mark Fisher. 2018. Self-ensembling for visual domain adaptation. In International Conference on Learning Representations. https://openreview.net/forum?id=rkpoTaxA-

work page 2018
[14]

Yaroslav Ganin and Victor Lempitsky. 2015. Unsupervised Domain Adaptation by Backpropagation. In Proceedings of the 32nd International Conference on Machine Learning (Proceedings of Machine Learning Research) , Francis Bach and David Blei (Eds.), Vol. 37. PMLR, 1180–1189. http://proceedings.mlr.press/v37/ganin15.html

work page 2015
[15]

Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette, Mario Marchand, and Victor Lempitsky. 2016. Domain-Adversarial Training of Neural Networks. Journal of Machine Learning Research 17, 59 (2016), 1–35. http://jmlr.org/papers/v17/15-239.html

work page 2016
[16]

Bastiaan Kleijn, Mengjie Zhang, David Balduzzi, and Wen Li

Muhammad Ghifary, W. Bastiaan Kleijn, Mengjie Zhang, David Balduzzi, and Wen Li. 2016. Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation. In Computer Vision – ECCV 2016 , Bastian Leibe, Jiri Matas, Nicu Sebe, and Max Welling (Eds.). Springer International Publishing, Cham, 597–613

work page 2016
[17]

Ian Goodfellow, Yoshua Bengio, and Aaron Courville. 2016. Deep learning. MIT press

work page 2016
[18]

Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative Adversarial Nets. In Advances in Neural Information Processing Systems 27 , Z. Ghahramani, M. Welling, C. Cortes, N. D. Lawrence, and K. Q. Weinberger (Eds.). Curran Asso- ciates, Inc., 2672–2680. http://papers....

work page 2014
[19]

Yves Grandvalet and Yoshua Bengio. 2005. Semi-supervised Learning by Entropy Minimization. In Advances in Neural Information Processing Systems 17 , L. K. Saul, Y. Weiss, and L. Bottou (Eds.). MIT Press, 529–536. http://papers.nips.cc/paper/ 2740-semi-supervised-learning-by-entropy-minimization.pdf

work page 2005
[20]

Guoliang Kang, Lu Jiang, Yi Yang, and Alexander G Hauptmann. 2019. Con- trastive Adaptation Network for Unsupervised Domain Adaptation.arXiv preprint arXiv:1901.00976 (2019)

work page internal anchor Pith review Pith/arXiv arXiv 2019
[21]

Abhishek Kumar, Prasanna Sattigeri, Kahini Wadhawan, Leonid Karlinsky, Roge- rio Feris, Bill Freeman, and Gregory Wornell. 2018. Co-regularized Alignment for Unsupervised Domain Adaptation. In Advances in Neural Information Processing Systems 31, S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett (Eds.). Curran Associates, I...

work page 2018
[22]

Samuli Laine and Timo Aila. 2017. Temporal Ensembling for Semi-Supervised Learning. In International Conference on Learning Representations . https: //openreview.net/forum?id=BJ6oOfqge

work page 2017
[23]

Yann LeCun, Corinna Cortes, and Christopher J.C. Burges. 1998. The MNIST database of handwritten digits. Retrieved August 16, 2018 from http://yann. lecun.com/exdb/mnist/

work page 1998
[24]

LeCun, O

Y. LeCun, O. Matan, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, L. D. Jacket, and H. S. Baird. 1990. Handwritten zip code recognition with multilayer networks. In [1990] Proceedings. 10th International Conference on Pattern Recognition, Vol. ii. 35–40 vol.2. https://doi.org/10.1109/ICPR.1990.119325

work page doi:10.1109/icpr.1990.119325 1990
[25]

Dong-Hyun Lee. 2013. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In Workshop on Challenges in Repre- sentation Learning, ICML, Vol. 3. 2

work page 2013
[26]

Yanghao Li, Naiyan Wang, Jianping Shi, Xiaodi Hou, and Jiaying Liu. 2018. Adap- tive Batch Normalization for practical domain adaptation. Pattern Recognition 80 (2018), 109 – 117. https://doi.org/10.1016/j.patcog.2018.03.005

work page doi:10.1016/j.patcog.2018.03.005 2018
[27]

Mingsheng Long, Yue Cao, Jianmin Wang, and Michael Jordan. 2015. Learning Transferable Features with Deep Adaptation Networks. InProceedings of the 32nd International Conference on Machine Learning (Proceedings of Machine Learning Research), Francis Bach and David Blei (Eds.), Vol. 37. PMLR, 97–105. http: //proceedings.mlr.press/v37/long15.html

work page 2015
[28]

Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan. 2018. Conditional Adversarial Domain Adaptation. In Advances in Neural Information Processing Systems 31, S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa- Bianchi, and R. Garnett (Eds.). Curran Associates, Inc., 1640–1650. http://papers. nips.cc/paper/7436-conditional-adversarial-...

work page 2018
[29]

Mingsheng Long, Han Zhu, Jianmin Wang, and Michael I. Jordan. 2017. Deep Transfer Learning with Joint Adaptation Networks. In Proceedings of the 34th International Conference on Machine Learning (Proceedings of Machine Learning Research), Doina Precup and Yee Whye Teh (Eds.), Vol. 70. PMLR, International Convention Centre, Sydney, Australia, 2208–2217. ht...

work page 2017
[30]

Miyato, S

T. Miyato, S. Maeda, S. Ishii, and M. Koyama. 2018. Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence (2018), 1–1. https: //doi.org/10.1109/TPAMI.2018.2858821

work page doi:10.1109/tpami.2018.2858821 2018
[31]

Boris Moiseev, Artem Konev, Alexander Chigorin, and Anton Konushin. 2013. Evaluation of Traffic Sign Recognition Methods Trained on Synthetically Gener- ated Data. InAdvanced Concepts for Intelligent Vision Systems, Jacques Blanc-Talon, Andrzej Kasinski, Wilfried Philips, Dan Popescu, and Paul Scheunders (Eds.). Springer International Publishing, Cham, 576–583

work page 2013
[32]

Pietro Morerio, Jacopo Cavazza, and Vittorio Murino. 2018. Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation. In Interna- tional Conference on Learning Representations . https://openreview.net/forum?id= rJWechg0Z

work page 2018
[33]

Pietro Morerio and Vittorio Murino. 2017. Correlation Alignment by Riemannian Metric for Domain Adaptation. arXiv preprint arXiv:1705.08180 (2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017
[34]

Yuval Netzer, Tao Wang, Adam Coates, Alessandro Bissacco, Bo Wu, and An- drew Y Ng. 2011. Reading digits in natural images with unsupervised feature learning. In NIPS workshop on deep learning and unsupervised feature learning , Vol. 2011. 5

work page 2011
[35]

Avital Oliver, Augustus Odena, Colin A Raffel, Ekin Dogus Cubuk, and Ian Good- fellow. 2018. Realistic Evaluation of Deep Semi-Supervised Learning Algorithms. In Advances in Neural Information Processing Systems 31 , S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett (Eds.). Curran Asso- ciates, Inc., 3235–3246. http://paper...

work page 2018
[36]

Sinno Jialin Pan and Qiang Yang. 2010. A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering 22, 10 (Oct 2010), 1345–1359. https://doi.org/10.1109/TKDE.2009.191

work page doi:10.1109/tkde.2009.191 2010
[37]

Sanjay Purushotham, Wilka Carvalho, Tanachat Nilanon, and Yan Liu. 2017. Vari- ational adversarial deep domain adaptation for health care time series analysis. In International Conference on Learning Representations . https://openreview.net/ forum?id=rk9eAFcxg

work page 2017
[38]

Rozantsev, M

A. Rozantsev, M. Salzmann, and P. Fua. 2019. Beyond Sharing Weights for Deep Domain Adaptation. IEEE Transactions on Pattern Analysis and Machine Intelligence 41, 4 (April 2019), 801–814. https://doi.org/10.1109/TPAMI.2018. 2814042

work page doi:10.1109/tpami.2018 2019
[39]

Carlucci, Tatiana Tommasi, and Barbara Caputo

Paolo Russo, Fabio M. Carlucci, Tatiana Tommasi, and Barbara Caputo. 2018. From Source to Target and Back: Symmetric Bi-Directional Adaptive GAN. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2018
[40]

Kate Saenko, Brian Kulis, Mario Fritz, and Trevor Darrell. 2010. Adapting Visual Category Models to New Domains. In Computer Vision – ECCV 2010 , Kostas Daniilidis, Petros Maragos, and Nikos Paragios (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 213–226

work page 2010
[41]

Kuniaki Saito, Yoshitaka Ushiku, and Tatsuya Harada. 2017. Asymmetric Tri- training for Unsupervised Domain Adaptation. In Proceedings of the 34th In- ternational Conference on Machine Learning (Proceedings of Machine Learning Research), Doina Precup and Yee Whye Teh (Eds.), Vol. 70. PMLR, 2988–2997. http://proceedings.mlr.press/v70/saito17a.html AdvML’19...

work page 2017
[42]

Kuniaki Saito, Yoshitaka Ushiku, Tatsuya Harada, and Kate Saenko. 2018. Adver- sarial Dropout Regularization. In International Conference on Learning Represen- tations. https://openreview.net/forum?id=HJIoJWZCZ

work page 2018
[43]

Castillo, and Rama Chel- lappa

Swami Sankaranarayanan, Yogesh Balaji, Carlos D. Castillo, and Rama Chel- lappa. 2018. Generate to Adapt: Aligning Domains Using Generative Adversarial Networks. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2018
[44]

Swami Sankaranarayanan, Yogesh Balaji, Arpit Jain, Ser Nam Lim, and Rama Chellappa. 2018. Learning From Synthetic Data: Addressing Domain Shift for Semantic Segmentation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2018
[45]

Ozan Sener, Hyun Oh Song, Ashutosh Saxena, and Silvio Savarese. 2016. Learn- ing Transferrable Representations for Unsupervised Domain Adaptation. In Advances in Neural Information Processing Systems 29 , D. D. Lee, M. Sugiyama, U. V. Luxburg, I. Guyon, and R. Garnett (Eds.). Curran Associates, Inc., 2110–

work page 2016
[46]

http://papers.nips.cc/paper/6360-learning-transferrable-representations- for-unsupervised-domain-adaptation.pdf

work page
[47]

Jian Shen, Yanru Qu, Weinan Zhang, and Yong Yu. 2018. Wasserstein Distance Guided Representation Learning for Domain Adaptation. In Thirty-Second AAAI Conference on Artificial Intelligence

work page 2018
[48]

Rui Shu, Hung Bui, Hirokazu Narui, and Stefano Ermon. 2018. A DIRT-T Ap- proach to Unsupervised Domain Adaptation. InInternational Conference on Learn- ing Representations. https://openreview.net/forum?id=H1q-TM-AW

work page 2018
[49]

Stallkamp, M

J. Stallkamp, M. Schlipsing, J. Salmen, and C. Igel. 2011. The German Traffic Sign Recognition Benchmark: A multi-class classification competition. In The 2011 International Joint Conference on Neural Networks . 1453–1460. https://doi.org/10. 1109/IJCNN.2011.6033395

work page arXiv 2011
[50]

Baochen Sun and Kate Saenko. 2016. Deep CORAL: Correlation Alignment for Deep Domain Adaptation. In Computer Vision – ECCV 2016 Workshops , Gang Hua and Hervé Jégou (Eds.). Springer International Publishing, Cham, 443–450

work page 2016
[51]

Antti Tarvainen and Harri Valpola. 2017. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. In Advances in Neural Information Processing Systems 30 , I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.). Curran Associates, Inc., 1195–1204. ht...

work page 2017
[52]

Eric Tzeng, Judy Hoffman, Trevor Darrell, and Kate Saenko. 2015. Simultaneous Deep Transfer Across Domains and Tasks. In The IEEE International Conference on Computer Vision (ICCV)

work page 2015
[53]

Eric Tzeng, Judy Hoffman, Kate Saenko, and Trevor Darrell. 2017. Adversarial Discriminative Domain Adaptation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2017
[54]

Jiawei Wang, Zhaoshui He, Chengjian Feng, Zhouping Zhu, Qinzhuang Lin, Jun Lv, and Shengli Xie. 2018. Domain Confusion with Self Ensembling for Unsupervised Adaptation. arXiv preprint arXiv:1810.04472 (2018)

work page arXiv 2018
[55]

Yifei Wang, Wen Li, Dengxin Dai, and Luc Van Gool. 2017. Deep Domain Adap- tation by Geodesic Distance Minimization. In The IEEE International Conference on Computer Vision (ICCV) Workshops

work page 2017
[56]

Kai-Ya Wei and Chiou-Ting Hsu. 2018. Generative Adversarial Guided Learning for Domain Adaptation. British Machine Vision Conference (2018)

work page 2018
[57]

Garrett Wilson and Diane J. Cook. 2019. A Survey of Unsupervised Deep Domain Adaptation. arXiv preprint arXiv:1812.02849 (2019)

work page arXiv 2019
[58]

Zhang, N

Y. Zhang, N. Wang, S. Cai, and L. Song. 2018. Unsupervised Domain Adaptation by Mapped Correlation Alignment. IEEE Access 6 (2018), 44698–44706. https: //doi.org/10.1109/ACCESS.2018.2865249

work page doi:10.1109/access.2018.2865249 2018
[59]

Han Zhao, Remi Tachet des Combes, Kun Zhang, and Geoffrey J Gordon. 2019. On Learning Invariant Representation for Domain Adaptation. arXiv preprint arXiv:1901.09453 (2019)

work page internal anchor Pith review Pith/arXiv arXiv 2019
[60]

Xiaojin Jerry Zhu. 2005. Semi-supervised learning literature survey . Technical Report. University of Wisconsin-Madison Department of Computer Sciences

work page 2005
[61]

Vijaya Kumar, and Jinsong Wang

Yang Zou, Zhiding Yu, B.V.K. Vijaya Kumar, and Jinsong Wang. 2018. Unsu- pervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training. In The European Conference on Computer Vision (ECCV)

work page 2018

[1] [1]

Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette, and Mario Marchand. 2014. Domain-adversarial neural networks. arXiv preprint arXiv:1412.4446 (2014)

work page internal anchor Pith review Pith/arXiv arXiv 2014

[2] [2]

Massih-Reza Amini and Patrick Gallinari. 2002. Semi-supervised logistic regres- sion. In ECAI. 390–394. Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence AdvML’19: Workshop on Adversarial Learning Methods for Machine Learning and Data Mining at KDD, August 5th, 2019, Anchorage, Alaska, USA

work page 2002

[3] [3]

Oscar Beijbom. 2012. Domain adaptations for computer vision applications.arXiv preprint arXiv:1211.4860 (2012)

work page internal anchor Pith review Pith/arXiv arXiv 2012

[4] [4]

Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman Vaughan. 2010. A theory of learning from different domains. Machine Learning 79, 1 (01 May 2010), 151–175. https://doi.org/10.1007/s10994- 009-5152-4

work page doi:10.1007/s10994- 2010

[5] [5]

Konstantinos Bousmalis, George Trigeorgis, Nathan Silberman, Dilip Krishnan, and Dumitru Erhan. 2016. Domain Separation Networks. In Advances in Neural Information Processing Systems 29, D. D. Lee, M. Sugiyama, U. V. Luxburg, I. Guyon, and R. Garnett (Eds.). Curran Associates, Inc., 343–351. http://papers.nips.cc/ paper/6254-domain-separation-networks.pdf

work page 2016

[6] [6]

Fabio Maria Carlucci, Lorenzo Porzi, Barbara Caputo, Elisa Ricci, and Samuel Rota Bulò. 2017. Autodial: Automatic domain alignment layers. In 2017 IEEE Interna- tional Conference on Computer Vision (ICCV) . IEEE, 5077–5085

work page 2017

[7] [7]

Olivier Chapelle and Alexander Zien. 2005. Semi-supervised classification by low density separation.. In AISTATS, Vol. 2005. Citeseer, 57–64

work page 2005

[8] [8]

Minmin Chen, Kilian Q Weinberger, and John Blitzer. 2011. Co-Training for Domain Adaptation. In Advances in Neural Information Processing Systems 24 , J. Shawe-Taylor, R. S. Zemel, P. L. Bartlett, F. Pereira, and K. Q. Weinberger (Eds.). Curran Associates, Inc., 2456–2464. http://papers.nips.cc/paper/4433-co- training-for-domain-adaptation.pdf

work page 2011

[9] [9]

Nicolas Courty, Rémi Flamary, Amaury Habrard, and Alain Rakotomamonjy. 2017. Joint distribution optimal transportation for domain adaptation. In Advances in Neural Information Processing Systems 30 , I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.). Curran Associates, Inc., 3730–3739. http://papers.nips.cc/...

work page 2017

[10] [10]

Bharath Bhushan Damodaran, Benjamin Kellenberger, Rémi Flamary, Devis Tuia, and Nicolas Courty. 2018. DeepJDOT: Deep Joint Distribution Optimal Transport for Unsupervised Domain Adaptation. In Computer Vision – ECCV 2018 , Vittorio Ferrari, Martial Hebert, Cristian Sminchisescu, and Yair Weiss (Eds.). Springer International Publishing, Cham, 467–483

work page 2018

[11] [11]

Debasmit Das and C. S. George Lee. 2018. Graph Matching and Pseudo-Label Guided Deep Unsupervised Domain Adaptation. InArtificial Neural Networks and Machine Learning – ICANN 2018 , Věra Kůrková, Yannis Manolopoulos, Barbara Hammer, Lazaros Iliadis, and Ilias Maglogiannis (Eds.). Springer International Publishing, Cham, 342–352

work page 2018

[12] [12]

Hal Daumé III. 2012. A course in machine learning. Publisher, ciml. info 5 (2012), 69

work page 2012

[13] [13]

Geoff French, Michal Mackiewicz, and Mark Fisher. 2018. Self-ensembling for visual domain adaptation. In International Conference on Learning Representations. https://openreview.net/forum?id=rkpoTaxA-

work page 2018

[14] [14]

Yaroslav Ganin and Victor Lempitsky. 2015. Unsupervised Domain Adaptation by Backpropagation. In Proceedings of the 32nd International Conference on Machine Learning (Proceedings of Machine Learning Research) , Francis Bach and David Blei (Eds.), Vol. 37. PMLR, 1180–1189. http://proceedings.mlr.press/v37/ganin15.html

work page 2015

[15] [15]

Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, François Laviolette, Mario Marchand, and Victor Lempitsky. 2016. Domain-Adversarial Training of Neural Networks. Journal of Machine Learning Research 17, 59 (2016), 1–35. http://jmlr.org/papers/v17/15-239.html

work page 2016

[16] [16]

Bastiaan Kleijn, Mengjie Zhang, David Balduzzi, and Wen Li

Muhammad Ghifary, W. Bastiaan Kleijn, Mengjie Zhang, David Balduzzi, and Wen Li. 2016. Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation. In Computer Vision – ECCV 2016 , Bastian Leibe, Jiri Matas, Nicu Sebe, and Max Welling (Eds.). Springer International Publishing, Cham, 597–613

work page 2016

[17] [17]

Ian Goodfellow, Yoshua Bengio, and Aaron Courville. 2016. Deep learning. MIT press

work page 2016

[18] [18]

Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative Adversarial Nets. In Advances in Neural Information Processing Systems 27 , Z. Ghahramani, M. Welling, C. Cortes, N. D. Lawrence, and K. Q. Weinberger (Eds.). Curran Asso- ciates, Inc., 2672–2680. http://papers....

work page 2014

[19] [19]

Yves Grandvalet and Yoshua Bengio. 2005. Semi-supervised Learning by Entropy Minimization. In Advances in Neural Information Processing Systems 17 , L. K. Saul, Y. Weiss, and L. Bottou (Eds.). MIT Press, 529–536. http://papers.nips.cc/paper/ 2740-semi-supervised-learning-by-entropy-minimization.pdf

work page 2005

[20] [20]

Guoliang Kang, Lu Jiang, Yi Yang, and Alexander G Hauptmann. 2019. Con- trastive Adaptation Network for Unsupervised Domain Adaptation.arXiv preprint arXiv:1901.00976 (2019)

work page internal anchor Pith review Pith/arXiv arXiv 2019

[21] [21]

Abhishek Kumar, Prasanna Sattigeri, Kahini Wadhawan, Leonid Karlinsky, Roge- rio Feris, Bill Freeman, and Gregory Wornell. 2018. Co-regularized Alignment for Unsupervised Domain Adaptation. In Advances in Neural Information Processing Systems 31, S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett (Eds.). Curran Associates, I...

work page 2018

[22] [22]

Samuli Laine and Timo Aila. 2017. Temporal Ensembling for Semi-Supervised Learning. In International Conference on Learning Representations . https: //openreview.net/forum?id=BJ6oOfqge

work page 2017

[23] [23]

Yann LeCun, Corinna Cortes, and Christopher J.C. Burges. 1998. The MNIST database of handwritten digits. Retrieved August 16, 2018 from http://yann. lecun.com/exdb/mnist/

work page 1998

[24] [24]

LeCun, O

Y. LeCun, O. Matan, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, L. D. Jacket, and H. S. Baird. 1990. Handwritten zip code recognition with multilayer networks. In [1990] Proceedings. 10th International Conference on Pattern Recognition, Vol. ii. 35–40 vol.2. https://doi.org/10.1109/ICPR.1990.119325

work page doi:10.1109/icpr.1990.119325 1990

[25] [25]

Dong-Hyun Lee. 2013. Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks. In Workshop on Challenges in Repre- sentation Learning, ICML, Vol. 3. 2

work page 2013

[26] [26]

Yanghao Li, Naiyan Wang, Jianping Shi, Xiaodi Hou, and Jiaying Liu. 2018. Adap- tive Batch Normalization for practical domain adaptation. Pattern Recognition 80 (2018), 109 – 117. https://doi.org/10.1016/j.patcog.2018.03.005

work page doi:10.1016/j.patcog.2018.03.005 2018

[27] [27]

Mingsheng Long, Yue Cao, Jianmin Wang, and Michael Jordan. 2015. Learning Transferable Features with Deep Adaptation Networks. InProceedings of the 32nd International Conference on Machine Learning (Proceedings of Machine Learning Research), Francis Bach and David Blei (Eds.), Vol. 37. PMLR, 97–105. http: //proceedings.mlr.press/v37/long15.html

work page 2015

[28] [28]

Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan. 2018. Conditional Adversarial Domain Adaptation. In Advances in Neural Information Processing Systems 31, S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa- Bianchi, and R. Garnett (Eds.). Curran Associates, Inc., 1640–1650. http://papers. nips.cc/paper/7436-conditional-adversarial-...

work page 2018

[29] [29]

Mingsheng Long, Han Zhu, Jianmin Wang, and Michael I. Jordan. 2017. Deep Transfer Learning with Joint Adaptation Networks. In Proceedings of the 34th International Conference on Machine Learning (Proceedings of Machine Learning Research), Doina Precup and Yee Whye Teh (Eds.), Vol. 70. PMLR, International Convention Centre, Sydney, Australia, 2208–2217. ht...

work page 2017

[30] [30]

Miyato, S

T. Miyato, S. Maeda, S. Ishii, and M. Koyama. 2018. Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence (2018), 1–1. https: //doi.org/10.1109/TPAMI.2018.2858821

work page doi:10.1109/tpami.2018.2858821 2018

[31] [31]

Boris Moiseev, Artem Konev, Alexander Chigorin, and Anton Konushin. 2013. Evaluation of Traffic Sign Recognition Methods Trained on Synthetically Gener- ated Data. InAdvanced Concepts for Intelligent Vision Systems, Jacques Blanc-Talon, Andrzej Kasinski, Wilfried Philips, Dan Popescu, and Paul Scheunders (Eds.). Springer International Publishing, Cham, 576–583

work page 2013

[32] [32]

Pietro Morerio, Jacopo Cavazza, and Vittorio Murino. 2018. Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation. In Interna- tional Conference on Learning Representations . https://openreview.net/forum?id= rJWechg0Z

work page 2018

[33] [33]

Pietro Morerio and Vittorio Murino. 2017. Correlation Alignment by Riemannian Metric for Domain Adaptation. arXiv preprint arXiv:1705.08180 (2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017

[34] [34]

Yuval Netzer, Tao Wang, Adam Coates, Alessandro Bissacco, Bo Wu, and An- drew Y Ng. 2011. Reading digits in natural images with unsupervised feature learning. In NIPS workshop on deep learning and unsupervised feature learning , Vol. 2011. 5

work page 2011

[35] [35]

Avital Oliver, Augustus Odena, Colin A Raffel, Ekin Dogus Cubuk, and Ian Good- fellow. 2018. Realistic Evaluation of Deep Semi-Supervised Learning Algorithms. In Advances in Neural Information Processing Systems 31 , S. Bengio, H. Wallach, H. Larochelle, K. Grauman, N. Cesa-Bianchi, and R. Garnett (Eds.). Curran Asso- ciates, Inc., 3235–3246. http://paper...

work page 2018

[36] [36]

Sinno Jialin Pan and Qiang Yang. 2010. A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering 22, 10 (Oct 2010), 1345–1359. https://doi.org/10.1109/TKDE.2009.191

work page doi:10.1109/tkde.2009.191 2010

[37] [37]

Sanjay Purushotham, Wilka Carvalho, Tanachat Nilanon, and Yan Liu. 2017. Vari- ational adversarial deep domain adaptation for health care time series analysis. In International Conference on Learning Representations . https://openreview.net/ forum?id=rk9eAFcxg

work page 2017

[38] [38]

Rozantsev, M

A. Rozantsev, M. Salzmann, and P. Fua. 2019. Beyond Sharing Weights for Deep Domain Adaptation. IEEE Transactions on Pattern Analysis and Machine Intelligence 41, 4 (April 2019), 801–814. https://doi.org/10.1109/TPAMI.2018. 2814042

work page doi:10.1109/tpami.2018 2019

[39] [39]

Carlucci, Tatiana Tommasi, and Barbara Caputo

Paolo Russo, Fabio M. Carlucci, Tatiana Tommasi, and Barbara Caputo. 2018. From Source to Target and Back: Symmetric Bi-Directional Adaptive GAN. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2018

[40] [40]

Kate Saenko, Brian Kulis, Mario Fritz, and Trevor Darrell. 2010. Adapting Visual Category Models to New Domains. In Computer Vision – ECCV 2010 , Kostas Daniilidis, Petros Maragos, and Nikos Paragios (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 213–226

work page 2010

[41] [41]

Kuniaki Saito, Yoshitaka Ushiku, and Tatsuya Harada. 2017. Asymmetric Tri- training for Unsupervised Domain Adaptation. In Proceedings of the 34th In- ternational Conference on Machine Learning (Proceedings of Machine Learning Research), Doina Precup and Yee Whye Teh (Eds.), Vol. 70. PMLR, 2988–2997. http://proceedings.mlr.press/v70/saito17a.html AdvML’19...

work page 2017

[42] [42]

Kuniaki Saito, Yoshitaka Ushiku, Tatsuya Harada, and Kate Saenko. 2018. Adver- sarial Dropout Regularization. In International Conference on Learning Represen- tations. https://openreview.net/forum?id=HJIoJWZCZ

work page 2018

[43] [43]

Castillo, and Rama Chel- lappa

Swami Sankaranarayanan, Yogesh Balaji, Carlos D. Castillo, and Rama Chel- lappa. 2018. Generate to Adapt: Aligning Domains Using Generative Adversarial Networks. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2018

[44] [44]

Swami Sankaranarayanan, Yogesh Balaji, Arpit Jain, Ser Nam Lim, and Rama Chellappa. 2018. Learning From Synthetic Data: Addressing Domain Shift for Semantic Segmentation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2018

[45] [45]

Ozan Sener, Hyun Oh Song, Ashutosh Saxena, and Silvio Savarese. 2016. Learn- ing Transferrable Representations for Unsupervised Domain Adaptation. In Advances in Neural Information Processing Systems 29 , D. D. Lee, M. Sugiyama, U. V. Luxburg, I. Guyon, and R. Garnett (Eds.). Curran Associates, Inc., 2110–

work page 2016

[46] [46]

http://papers.nips.cc/paper/6360-learning-transferrable-representations- for-unsupervised-domain-adaptation.pdf

work page

[47] [47]

Jian Shen, Yanru Qu, Weinan Zhang, and Yong Yu. 2018. Wasserstein Distance Guided Representation Learning for Domain Adaptation. In Thirty-Second AAAI Conference on Artificial Intelligence

work page 2018

[48] [48]

Rui Shu, Hung Bui, Hirokazu Narui, and Stefano Ermon. 2018. A DIRT-T Ap- proach to Unsupervised Domain Adaptation. InInternational Conference on Learn- ing Representations. https://openreview.net/forum?id=H1q-TM-AW

work page 2018

[49] [49]

Stallkamp, M

J. Stallkamp, M. Schlipsing, J. Salmen, and C. Igel. 2011. The German Traffic Sign Recognition Benchmark: A multi-class classification competition. In The 2011 International Joint Conference on Neural Networks . 1453–1460. https://doi.org/10. 1109/IJCNN.2011.6033395

work page arXiv 2011

[50] [50]

Baochen Sun and Kate Saenko. 2016. Deep CORAL: Correlation Alignment for Deep Domain Adaptation. In Computer Vision – ECCV 2016 Workshops , Gang Hua and Hervé Jégou (Eds.). Springer International Publishing, Cham, 443–450

work page 2016

[51] [51]

Antti Tarvainen and Harri Valpola. 2017. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. In Advances in Neural Information Processing Systems 30 , I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.). Curran Associates, Inc., 1195–1204. ht...

work page 2017

[52] [52]

Eric Tzeng, Judy Hoffman, Trevor Darrell, and Kate Saenko. 2015. Simultaneous Deep Transfer Across Domains and Tasks. In The IEEE International Conference on Computer Vision (ICCV)

work page 2015

[53] [53]

Eric Tzeng, Judy Hoffman, Kate Saenko, and Trevor Darrell. 2017. Adversarial Discriminative Domain Adaptation. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

work page 2017

[54] [54]

Jiawei Wang, Zhaoshui He, Chengjian Feng, Zhouping Zhu, Qinzhuang Lin, Jun Lv, and Shengli Xie. 2018. Domain Confusion with Self Ensembling for Unsupervised Adaptation. arXiv preprint arXiv:1810.04472 (2018)

work page arXiv 2018

[55] [55]

Yifei Wang, Wen Li, Dengxin Dai, and Luc Van Gool. 2017. Deep Domain Adap- tation by Geodesic Distance Minimization. In The IEEE International Conference on Computer Vision (ICCV) Workshops

work page 2017

[56] [56]

Kai-Ya Wei and Chiou-Ting Hsu. 2018. Generative Adversarial Guided Learning for Domain Adaptation. British Machine Vision Conference (2018)

work page 2018

[57] [57]

Garrett Wilson and Diane J. Cook. 2019. A Survey of Unsupervised Deep Domain Adaptation. arXiv preprint arXiv:1812.02849 (2019)

work page arXiv 2019

[58] [58]

Zhang, N

Y. Zhang, N. Wang, S. Cai, and L. Song. 2018. Unsupervised Domain Adaptation by Mapped Correlation Alignment. IEEE Access 6 (2018), 44698–44706. https: //doi.org/10.1109/ACCESS.2018.2865249

work page doi:10.1109/access.2018.2865249 2018

[59] [59]

Han Zhao, Remi Tachet des Combes, Kun Zhang, and Geoffrey J Gordon. 2019. On Learning Invariant Representation for Domain Adaptation. arXiv preprint arXiv:1901.09453 (2019)

work page internal anchor Pith review Pith/arXiv arXiv 2019

[60] [60]

Xiaojin Jerry Zhu. 2005. Semi-supervised learning literature survey . Technical Report. University of Wisconsin-Madison Department of Computer Sciences

work page 2005

[61] [61]

Vijaya Kumar, and Jinsong Wang

Yang Zou, Zhiding Yu, B.V.K. Vijaya Kumar, and Jinsong Wang. 2018. Unsu- pervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training. In The European Conference on Computer Vision (ECCV)

work page 2018