Uncertainty-Guided Mixup for Semi-Supervised Domain Adaptation without Source Data

Present domain adaptation methods usually perform explicit representation alignment by simultaneously accessing the source data and target data. However, the source data are not always available due to the privacy preserving consideration or bandwidth limitation. Source-free domain adaptation aims to solve the above problem by performing domain adaptation without accessing the source data. The adaptation paradigm is receiving more and more attention in recent years, and multiple works have been proposed for unsupervised source-free domain adaptation. However, without utilizing any supervised signal and source data at the adaptation stage, the optimization of the target model is unstable and fragile. To alleviate the problem, we focus on semi-supervised domain adaptation under source-free setting. More specifically, we propose uncertainty-guided Mixup to reduce the representation’s intra-domain discrepancy and perform inter-domain alignment without directly accessing the source data. We conduct extensive semi-supervised domain adaptation experiments on various datasets. Our method outperforms the recent semi-supervised baselines and the unsupervised variant also achieves competitive performance. The experiment codes will be released in the future.

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

[2]  Harri Valpola,et al.  Weight-averaged consistency targets improve semi-supervised deep learning results , 2017, ArXiv.

[3]  Xifeng Yan,et al.  Adaptive-Step Graph Meta-Learner for Few-Shot Graph Classification , 2020, CIKM.

[4]  David Berthelot,et al.  MixMatch: A Holistic Approach to Semi-Supervised Learning , 2019, NeurIPS.

[5]  Peter L. Bartlett,et al.  Rademacher and Gaussian Complexities: Risk Bounds and Structural Results , 2003, J. Mach. Learn. Res..

[6]  Kate Saenko,et al.  Adversarial Dropout Regularization , 2017, ICLR.

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

[8]  Victor S. Lempitsky,et al.  Unsupervised Domain Adaptation by Backpropagation , 2014, ICML.

[9]  Trevor Darrell,et al.  Fully Test-time Adaptation by Entropy Minimization , 2020, ArXiv.

[10]  Zhun Deng,et al.  How Does Mixup Help With Robustness and Generalization? , 2020, ArXiv.

[11]  Hongyi Zhang,et al.  mixup: Beyond Empirical Risk Minimization , 2017, ICLR.

[12]  Avishek Saha,et al.  Co-regularization Based Semi-supervised Domain Adaptation , 2010, NIPS.

[13]  Sergey Levine,et al.  Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks , 2017, ICML.

[14]  Kate Saenko,et al.  Surprisingly Simple Semi-Supervised Domain Adaptation with Pretraining and Consistency , 2021, BMVC.

[15]  Shin Ishii,et al.  Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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

[17]  Min Xiao,et al.  Semi-Supervised Kernel Matching for Domain Adaptation , 2012, AAAI.

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

[19]  Bo Wang,et al.  Moment Matching for Multi-Source Domain Adaptation , 2018, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

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

[21]  Changick Kim,et al.  Attract, Perturb, and Explore: Learning a Feature Alignment Network for Semi-supervised Domain Adaptation , 2020, ECCV.

[22]  Kate Saenko,et al.  Federated Adversarial Domain Adaptation , 2020, ICLR.

[23]  Alex Kendall,et al.  What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? , 2017, NIPS.

[24]  Zhedong Zheng,et al.  Rectifying Pseudo Label Learning via Uncertainty Estimation for Domain Adaptive Semantic Segmentation , 2021, Int. J. Comput. Vis..

[25]  Hau-San Wong,et al.  Model Adaptation: Unsupervised Domain Adaptation Without Source Data , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[27]  Jiashi Feng,et al.  Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation , 2020, ICML.

[28]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[29]  Bingbing Ni,et al.  Adversarial Domain Adaptation with Domain Mixup , 2019, AAAI.

[30]  Julien Cornebise,et al.  Weight Uncertainty in Neural Network , 2015, ICML.

[31]  Yunfeng Shao,et al.  Bidirectional Adversarial Training for Semi-Supervised Domain Adaptation , 2020, IJCAI.

[32]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[33]  Gabriela Csurka,et al.  Domain Adaptation in the Absence of Source Domain Data , 2016, KDD.

[34]  Timothy Hospedales,et al.  Online Meta-Learning for Multi-Source and Semi-Supervised Domain Adaptation , 2020, ECCV.

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

[36]  Mubarak Shah,et al.  In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning , 2021, ICLR.

[37]  A. Müller.  Integral Probability Metrics and Their Generating Classes of Functions , 1997, Advances in Applied Probability.

[38]  Yoshua Bengio,et al.  Semi-supervised Learning by Entropy Minimization , 2004, CAP.

[39]  Minghao Chen,et al.  KD3A: Unsupervised Multi-Source Decentralized Domain Adaptation via Knowledge Distillation , 2021, ICML.