Bridging the Theoretical Bound and Deep Algorithms for Open Set Domain Adaptation

In the unsupervised open set domain adaptation (UOSDA), the target domain contains unknown classes that are not observed in the source domain. Researchers in this area aim to train a classifier to accurately: 1) recognize unknown target data (data with unknown classes) and, 2) classify other target data. To achieve this aim, a previous study has proven an upper bound of the target-domain risk, and the open set difference, as an important term in the upper bound, is used to measure the risk on unknown target data. By minimizing the upper bound, a shallow classifier can be trained to achieve the aim. However, if the classifier is very flexible (e.g., deep neural networks (DNNs)), the open set difference will converge to a negative value when minimizing the upper bound, which causes an issue where most target data are recognized as unknown data. To address this issue, we propose a new upper bound of target-domain risk for UOSDA, which includes four terms: source-domain risk, $\epsilon$-open set difference ($\Delta_\epsilon$), a distributional discrepancy between domains, and a constant. Compared to the open set difference, $\Delta_\epsilon$ is more robust against the issue when it is being minimized, and thus we are able to use very flexible classifiers (i.e., DNNs). Then, we propose a new principle-guided deep UOSDA method that trains DNNs via minimizing the new upper bound. Specifically, source-domain risk and $\Delta_\epsilon$ are minimized by gradient descent, and the distributional discrepancy is minimized via a novel open-set conditional adversarial training strategy. Finally, compared to existing shallow and deep UOSDA methods, our method shows the state-of-the-art performance on several benchmark datasets, including digit recognition (MNIST, SVHN, USPS), object recognition (Office-31, Office-Home), and face recognition (PIE).

[1]  Yi Yang,et al.  Attract or Distract: Exploit the Margin of Open Set , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[2]  Feng Liu,et al.  Unsupervised Domain Adaptation with Sphere Retracting Transformation , 2019, 2019 International Joint Conference on Neural Networks (IJCNN).

[3]  Qian Zhang,et al.  A cross-domain recommender system with consistent information transfer , 2017, Decis. Support Syst..

[4]  Ricardo da Silva Torres,et al.  Semi-supervised transfer subspace for domain adaptation , 2018, Pattern Recognit..

[5]  Sethuraman Panchanathan,et al.  Deep Hashing Network for Unsupervised Domain Adaptation , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Trevor Darrell,et al.  Adapting Visual Category Models to New Domains , 2010, ECCV.

[7]  Feng Liu,et al.  Open Set Domain Adaptation: Theoretical Bound and Algorithm , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[8]  Witold Pedrycz,et al.  Granular Fuzzy Regression Domain Adaptation in Takagi–Sugeno Fuzzy Models , 2018, IEEE Transactions on Fuzzy Systems.

[9]  Yuchen Zhang,et al.  Bridging Theory and Algorithm for Domain Adaptation , 2019, ICML.

[10]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[11]  Yishay Mansour,et al.  Domain Adaptation: Learning Bounds and Algorithms , 2009, COLT.

[12]  Andrew Y. Ng,et al.  Reading Digits in Natural Images with Unsupervised Feature Learning , 2011 .

[13]  Trevor Darrell,et al.  DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition , 2013, ICML.

[14]  Donald A. Adjeroh,et al.  Unified Deep Supervised Domain Adaptation and Generalization , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[15]  Juergen Gall,et al.  Open Set Domain Adaptation , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[16]  Peng Hao,et al.  Transfer learning using computational intelligence: A survey , 2015, Knowl. Based Syst..

[17]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[18]  Witold Pedrycz,et al.  Fuzzy Rule-Based Domain Adaptation in Homogeneous and Heterogeneous Spaces , 2019, IEEE Transactions on Fuzzy Systems.

[19]  Terrance E. Boult,et al.  Towards Open Set Deep Networks , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Jie Lu,et al.  Heterogeneous unsupervised domain adaptation based on fuzzy feature fusion , 2017, 2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE).

[21]  Stefano Ermon,et al.  A DIRT-T Approach to Unsupervised Domain Adaptation , 2018, ICLR.

[22]  Philip S. Yu,et al.  Visual Domain Adaptation with Manifold Embedded Distribution Alignment , 2018, ACM Multimedia.

[23]  Jianmin Wang,et al.  Partial Transfer Learning with Selective Adversarial Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[24]  Hong Liu,et al.  Separate to Adapt: Open Set Domain Adaptation via Progressive Separation , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Feng Liu,et al.  Heterogeneous Domain Adaptation: An Unsupervised Approach , 2020, IEEE Transactions on Neural Networks and Learning Systems.

[26]  Feng Liu,et al.  Fuzzy Transfer Learning Using an Infinite Gaussian Mixture Model and Active Learning , 2019, IEEE Transactions on Fuzzy Systems.

[27]  Jian Shen,et al.  Wasserstein Distance Guided Representation Learning for Domain Adaptation , 2017, AAAI.

[28]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[29]  Philip S. Yu,et al.  Transfer Feature Learning with Joint Distribution Adaptation , 2013, 2013 IEEE International Conference on Computer Vision.

[30]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[31]  Michael I. Jordan,et al.  Conditional Adversarial Domain Adaptation , 2017, NeurIPS.

[32]  Jindong Wang,et al.  Easy Transfer Learning By Exploiting Intra-Domain Structures , 2019, 2019 IEEE International Conference on Multimedia and Expo (ICME).

[33]  Ivor W. Tsang,et al.  Domain Adaptation via Transfer Component Analysis , 2009, IEEE Transactions on Neural Networks.

[34]  Ricardo da Silva Torres,et al.  Nearest neighbors distance ratio open-set classifier , 2016, Machine Learning.

[35]  Kate Saenko,et al.  Return of Frustratingly Easy Domain Adaptation , 2015, AAAI.

[36]  John K. Tsotsos,et al.  PIE: A Large-Scale Dataset and Models for Pedestrian Intention Estimation and Trajectory Prediction , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[37]  Rama Chellappa,et al.  Domain adaptation for object recognition: An unsupervised approach , 2011, 2011 International Conference on Computer Vision.

[38]  Trevor Darrell,et al.  Semi-Supervised Domain Adaptation via Minimax Entropy , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[39]  Jonathan J. Hull,et al.  A Database for Handwritten Text Recognition Research , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[40]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[41]  Shiguang Shan,et al.  Domain Adaptation for Face Recognition: Targetize Source Domain Bridged by Common Subspace , 2013, International Journal of Computer Vision.

[42]  Michael I. Jordan,et al.  Universal Domain Adaptation , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[43]  Terrance E. Boult,et al.  Multi-class Open Set Recognition Using Probability of Inclusion , 2014, ECCV.

[44]  Alexander J. Smola,et al.  Hilbert space embeddings of conditional distributions with applications to dynamical systems , 2009, ICML '09.

[45]  Bernhard Schölkopf,et al.  A Kernel Two-Sample Test , 2012, J. Mach. Learn. Res..

[46]  Michael I. Jordan,et al.  Learning Transferable Features with Deep Adaptation Networks , 2015, ICML.

[47]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[48]  Koby Crammer,et al.  Analysis of Representations for Domain Adaptation , 2006, NIPS.

[49]  Tatsuya Harada,et al.  Open Set Domain Adaptation by Backpropagation , 2018, ECCV.