ABSTRACT
The task of domain generalization is to learn a classification model from multiple source domains and generalize it to unknown target domains. The key to domain generalization is learning discriminative domain-invariant features. Invariant representations are achieved using adversarial domain generalization as one of the primary techniques. For example, generative adversarial networks have been widely used, but suffer from the problem of low intra-class diversity, which can lead to poor generalization ability. To address this issue, we propose a new method called auxiliary classifier in adversarial domain generalization (CloCls). CloCls improve the diversity of the source domain by introducing auxiliary classifier. Combining typical task-related losses, e.g., cross-entropy loss for classification and adversarial loss for domain discrimination, our overall goal is to guarantee the learning of condition-invariant features for all source domains while increasing the diversity of source domains. Further, inspired by smoothing optima have improved generalization for supervised learning tasks like classification. We leverage that converging to a smooth minima with respect task loss stabilizes the adversarial training leading to better performance on unseen target domain which can effectively enhances the performance of domain adversarial methods. We have conducted extensive image classification experiments on benchmark datasets in domain generalization, and our model exhibits sufficient generalization ability and outperforms state-of-the-art DG methods.
- Yogesh Balaji, Swami Sankaranarayanan, and Rama Chellappa. 2018. Metareg: Towards domain generalization using meta-regularization. NeurIPS, Vol. 31 (2018).Google Scholar
- 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, Vol. 79, 1 (2010), 151--175.Google Scholar
- Shai Ben-David, John Blitzer, Koby Crammer, and Fernando Pereira. 2006. Analysis of representations for domain adaptation. Advances in neural information processing systems, Vol. 19 (2006).Google Scholar
- Guanyu Cai, Yuqin Wang, Lianghua He, and MengChu Zhou. 2019. Unsupervised domain adaptation with adversarial residual transform networks. IEEE transactions on neural networks and learning systems (TNNLS), Vol. 31, 8 (2019), 3073--3086.Google Scholar
- Fabio M Carlucci, Antonio D'Innocente, Silvia Bucci, Barbara Caputo, and Tatiana Tommasi. 2019a. Domain generalization by solving jigsaw puzzles. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2229--2238.Google ScholarCross Ref
- Fabio Maria Carlucci, Antonio D'Innocente, S. Bucci, B. Caputo, and T. Tommasi. 2019b. Domain Generalization by Solving Jigsaw Puzzles. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2019), 2224--2233.Google Scholar
- Junbum Cha, Sanghyuk Chun, Kyungjae Lee, Han-Cheol Cho, Seunghyun Park, Yunsung Lee, and Sungrae Park. 2021. SWAD: Domain Generalization by Seeking Flat Minima. arXiv preprint arXiv:2102.08604 (2021).Google Scholar
- Prithvijit Chattopadhyay, Yogesh Balaji, and Judy Hoffman. 2020. Learning to balance specificity and invariance for in and out of domain generalization. In ECCV. 301--318.Google Scholar
- Chaoqi Chen, Jiongcheng Li, Xiaoguang Han, Xiaoqing Liu, and Yizhou Yu. 2022. Compound Domain Generalization via Meta-Knowledge Encoding. In CVPR. 7119--7129.Google Scholar
- Gintare Karolina Dziugaite and Daniel M Roy. 2017. Computing nonvacuous generalization bounds for deep (stochastic) neural networks with many more parameters than training data. arXiv preprint arXiv:1703.11008 (2017).Google Scholar
- Pierre Foret, Ariel Kleiner, Hossein Mobahi, and Behnam Neyshabur. 2021. Sharpness-aware Minimization for Efficiently Improving Generalization. In International Conference on Learning Representations. https://openreview.net/forum?id=6Tm1mposlrMGoogle Scholar
- Muhammad Ghifary, W Bastiaan Kleijn, Mengjie Zhang, and David Balduzzi. 2015. Domain generalization for object recognition with multi-task autoencoders. In ICCV. 2551--2559.Google Scholar
- Ishaan Gulrajani and David Lopez-Paz. 2021. In search of lost domain generalization. ICLR (2021).Google Scholar
- Haowei He, Gao Huang, and Yang Yuan. 2019. Asymmetric valleys: beyond sharp and flat local minima. In Proceedings of the 33rd International Conference on Neural Information Processing Systems. 2553--2564.Google Scholar
- Sepp Hochreiter and Jürgen Schmidhuber. 1994. Simplifying neural nets by discovering flat minima. Advances in neural information processing systems, Vol. 7 (1994).Google Scholar
- Sepp Hochreiter and Jürgen Schmidhuber. 1997. Flat minima. Neural computation, Vol. 9, 1 (1997), 1--42.Google Scholar
- Liang Hou, Qi Cao, Huawei Shen, Siyuan Pan, Xiaoshuang Li, and Xueqi Cheng. 2022. Conditional gans with auxiliary discriminative classifier. In International Conference on Machine Learning. PMLR, 8888--8902.Google Scholar
- Zeyi Huang, Haohan Wang, Eric P. Xing, and Dong Huang. 2020a. Self-challenging Improves Cross-Domain Generalization. In European Conference on Computer Vision (ECCV). 124--140.Google Scholar
- Zeyi Huang, Haohan Wang, Eric P Xing, and Dong Huang. 2020b. Self-challenging improves cross-domain generalization. In ECCV. 124--140.Google Scholar
- Nitish Shirish Keskar, Jorge Nocedal, Ping Tak Peter Tang, Dheevatsa Mudigere, and Mikhail Smelyanskiy. 2017. On large-batch training for deep learning: Generalization gap and sharp minima. In 5th International Conference on Learning Representations, ICLR 2017.Google Scholar
- Daehee Kim, Youngjun Yoo, Seunghyun Park, Jinkyu Kim, and Jaekoo Lee. 2021. Selfreg: Self-supervised contrastive regularization for domain generalization. In ICCV. 9619--9628.Google Scholar
- Jungmin Kwon, Jeongseop Kim, Hyunseo Park, and In Kwon Choi. 2021. Asam: Adaptive sharpness-aware minimization for scale-invariant learning of deep neural networks. In International Conference on Machine Learning. PMLR, 5905--5914.Google Scholar
- Da Li, Yongxin Yang, Yi-Zhe Song, and Timothy M Hospedales. 2017. Deeper, broader and artier domain generalization. In Proceedings of the IEEE international conference on computer vision. 5542--5550.Google ScholarCross Ref
- Da Li, Yongxin Yang, Yi-Zhe Song, and Timothy M Hospedales. 2018d. Learning to generalize: Meta-learning for domain generalization. In AAAI.Google Scholar
- Haoliang Li, Sinno Jialin Pan, Shiqi Wang, and Alex C Kot. 2018b. Domain generalization with adversarial feature learning. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 5400--5409.Google ScholarCross Ref
- Haoliang Li, Yufei Wang, Renjie Wan, Shiqi Wang, Tie-Qiang Li, and Alex C. Kot. 2020. Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularization. In Advances in Neural Information Processing Systems (NeurIPS).Google Scholar
- Ya Li, Mingming Gong, Xinmei Tian, Tongliang Liu, and Dacheng Tao. 2018a. Domain generalization via conditional invariant representations. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 32.Google ScholarCross Ref
- Ya Li, Xinmei Tian, Mingming Gong, Yajing Liu, Tongliang Liu, Kun Zhang, and Dacheng Tao. 2018c. Deep domain generalization via conditional invariant adversarial networks. In Proceedings of the European Conference on Computer Vision (ECCV). 624--639.Google ScholarDigital Library
- Mingsheng Long, Yue Cao, Jianmin Wang, and Michael Jordan. 2015. Learning transferable features with deep adaptation networks. In International conference on machine learning. PMLR, 97--105.Google Scholar
- Ao Ma, Jingjing Li, Ke Lu, Lei Zhu, and Heng Tao Shen. 2021. Adversarial Entropy Optimization for Unsupervised Domain Adaptation. IEEE Transactions on Neural Networks and Learning Systems (TNNLS) (2021).Google Scholar
- Divyat Mahajan, Shruti Tople, and Amit Sharma. 2020. Domain generalization using causal matching. arXiv preprint arXiv:2006.07500 (2020).Google Scholar
- T. Matsuura and T. Harada. 2020. Domain Generalization Using a Mixture of Multiple Latent Domains. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI).Google Scholar
- Saeid Motiian, Marco Piccirilli, Donald A Adjeroh, and Gianfranco Doretto. 2017. Unified deep supervised domain adaptation and generalization. In ICCV. 5715--5725.Google Scholar
- Krikamol Muandet, David Balduzzi, and Bernhard Schölkopf. 2013. Domain generalization via invariant feature representation. In International Conference on Machine Learning. PMLR, 10--18.Google Scholar
- Hyeonseob Nam, HyunJae Lee, Jongchan Park, Wonjun Yoon, and Donggeun Yoo. 2021. Reducing domain gap by reducing style bias. In CVPR. 8690--8699.Google Scholar
- Sinno Jialin Pan, Ivor W Tsang, James T Kwok, and Qiang Yang. 2010. Domain adaptation via transfer component analysis. IEEE transactions on neural networks, Vol. 22, 2 (2010), 199--210.Google Scholar
- Xingchao Peng, Qinxun Bai, Xide Xia, Zijun Huang, Kate Saenko, and Bo Wang. 2019. Moment matching for multi-source domain adaptation. In Proceedings of the IEEE International Conference on Computer Vision. 1406--1415.Google ScholarCross Ref
- Vihari Piratla, Praneeth Netrapalli, and Sunita Sarawagi. 2020. Efficient Domain Generalization via Common-Specific Low-Rank Decomposition. In International Conference on Machine Learning (ICML).Google Scholar
- Fengchun Qiao, Long Zhao, and Xi Peng. 2020. Learning to learn single domain generalization. In CVPR. 12556--12565.Google Scholar
- Harsh Rangwani, Sumukh K Aithal, Mayank Mishra, Arihant Jain, and Venkatesh Babu Radhakrishnan. 2022. A closer look at smoothness in domain adversarial training. In International Conference on Machine Learning. PMLR, 18378--18399.Google Scholar
- Seonguk Seo, Yumin Suh, D. Kim, Jongwoo Han, and B. Han. 2020. Learning to Optimize Domain Specific Normalization for Domain Generalization. In European Conference on Computer Vision (ECCV).Google Scholar
- Shiv Shankar, Vihari Piratla, Soumen Chakrabarti, Siddhartha Chaudhuri, Preethi Jyothi, and Sunita Sarawagi. 2018. Generalizing across domains via cross-gradient training. arXiv preprint arXiv:1804.10745 (2018).Google Scholar
- Zhishu Sun, Zhifeng Shen, Luojun Lin, Yuanlong Yu, Zhifeng Yang, Shicai Yang, and Weijie Chen. 2022. Dynamic Domain Generalization. IJCAI (2022).Google Scholar
- Hemanth Venkateswara, Jose Eusebio, Shayok Chakraborty, and Sethuraman Panchanathan. 2017. Deep Hashing Network for Unsupervised Domain Adaptation. In (IEEE) Conference on Computer Vision and Pattern Recognition (CVPR).Google Scholar
- Riccardo Volpi, Hongseok Namkoong, Ozan Sener, John C Duchi, Vittorio Murino, and Silvio Savarese. 2018. Generalizing to unseen domains via adversarial data augmentation. Advances in neural information processing systems, Vol. 31 (2018).Google Scholar
- Mengzhu Wang, Jianlong Yuan, Qi Qian, Zhibin Wang, and Hao Li. 2022. Semantic Data Augmentation based Distance Metric Learning for Domain Generalization. In Proceedings of the 30th ACM International Conference on Multimedia. 3214--3223.Google ScholarDigital Library
- Shujun Wang, Lequan Yu, Caizi Li, Chi-Wing Fu, and Pheng-Ann Heng. 2020. Learning from extrinsic and intrinsic supervisions for domain generalization. In ECCV. 159--176.Google Scholar
- Zengmao Wang, Bo Du, and Yuhong Guo. 2019. Domain adaptation with neural embedding matching. IEEE transactions on neural networks and learning systems (TNNLS), Vol. 31, 7 (2019), 2387--2397.Google Scholar
- Xiaofu Wu, Suofei Zhang, Quan Zhou, Zhen Yang, Chunming Zhao, and Longin Jan Latecki. 2021. Entropy Minimization Versus Diversity Maximization for Domain Adaptation. IEEE Transactions on Neural Networks and Learning Systems (TNNLS) (2021).Google Scholar
- Qinwei Xu, Ruipeng Zhang, Ya Zhang, Yanfeng Wang, and Qi Tian. 2021. A fourier-based framework for domain generalization. In CVPR. 14383--14392.Google Scholar
- Fu-En Yang, Yuan-Chia Cheng, Zu-Yun Shiau, and Yu-Chiang Frank Wang. 2021. Adversarial teacher-student representation learning for domain generalization. Advances in Neural Information Processing Systems (2021), 19448--19460.Google Scholar
- Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. 2017. mixup: Beyond empirical risk minimization. ICLR (2017).Google Scholar
- Lei Zhang, Jingru Fu, Shanshan Wang, David Zhang, Zhaoyang Dong, and CL Philip Chen. 2019. Guide subspace learning for unsupervised domain adaptation. IEEE transactions on neural networks and learning systems (TNNLS), Vol. 31, 9 (2019), 3374--3388.Google Scholar
- Shanshan Zhao, Mingming Gong, Tongliang Liu, Huan Fu, and Dacheng Tao. 2020a. Domain generalization via entropy regularization. Advances in Neural Information Processing Systems, Vol. 33 (2020), 16096--16107.Google Scholar
- Shanshan Zhao, M. Gong, T. Liu, H. Fu, and Dacheng Tao. 2020b. Domain Generalization via Entropy Regularization. In Advances in Neural Information Processing Systems (NeurIPS).Google Scholar
- Kaiyang Zhou, Yongxin Yang, Timothy Hospedales, and Tao Xiang. 2020a. Deep domain-adversarial image generation for domain generalisation. In Proceedings of the AAAI Conference on Artificial Intelligence. 13025--13032.Google ScholarCross Ref
- Kaiyang Zhou, Yongxin Yang, Timothy Hospedales, and Tao Xiang. 2020b. Learning to generate novel domains for domain generalization. In ECCV. 561--578.Google Scholar
- Kaiyang Zhou, Yongxin Yang, Yu Qiao, and Tao Xiang. 2021. Domain generalization with mixstyle. ICLR (2021).Google Scholar
Index Terms
- A Closer Look at Classifier in Adversarial Domain Generalization
Recommendations
Decomposed adversarial domain generalization
AbstractWe tackle the problem of generalizing a predictor trained on a set of source domains to an unseen target domain, where the source and target domains are different but related to one another, i.e., the domain generalization problem. ...
Category-Stitch Learning for Union Domain Generalization
Domain generalization aims at generalizing the network trained on multiple domains to unknown but related domains. Under the assumption that different domains share the same classes, previous works can build relationships across domains. However, in ...
Cross-domain Ensemble Distillation for Domain Generalization
Computer Vision – ECCV 2022AbstractDomain generalization is the task of learning models that generalize to unseen target domains. We propose a simple yet effective method for domain generalization, named cross-domain ensemble distillation (XDED), that learns domain-invariant ...
Comments