{"type":"ScholarlyArticle","title":"Dynamic Domain Generalization","authors":[{"name":"Zhishu Sun","slug":"zhishu-sun","given":"Zhishu","family":"Sun","affiliation":"College of Computer and Data Science , Fuzhou University , Fuzhou , China"},{"name":"Zhifeng Shen","slug":"zhifeng-shen","given":"Zhifeng","family":"Shen","affiliation":"College of Computer and Data Science , Fuzhou University , Fuzhou , China"},{"name":"Luojun Lin","slug":"luojun-lin","given":"Luojun","family":"Lin","affiliation":"College of Computer and Data Science , Fuzhou University , Fuzhou , China"},{"name":"Yuanlong Yu","slug":"yuanlong-yu","given":"Yuanlong","family":"Yu","affiliation":"College of Computer and Data Science , Fuzhou University , Fuzhou , China"},{"name":"Zhifeng Yang","slug":"zhifeng-yang","given":"Zhifeng","family":"Yang","affiliation":"College of Computer and Data Science , Fuzhou University , Fuzhou , China"},{"name":"Shicai Yang","slug":"shicai-yang","given":"Shicai","family":"Yang","affiliation":"Hikvision Research Institute , Hangzhou , China"},{"name":"Weijie Chen","slug":"weijie-chen","given":"Weijie","family":"Chen","affiliation":"Hikvision Research Institute , Hangzhou , China"}],"abstract":"Domain generalization (DG) is a fundamental yet very challenging research topic in machine learning. The existing arts mainly focus on learning domain-invariant features with limited source domains in a static model. Unfortunately, there is a lack of training-free mechanism to adjust the model when generalized to the agnostic target domains. To tackle this problem, we develop a brand-new DG variant, namely Dynamic Domain Generalization (DDG), in which the model learns to twist the network parameters to adapt to the data from different domains. Specifically, we leverage a meta-adjuster to twist the network parameters based on the static model with respect to different data from different domains. In this way, the static model is optimized to learn domain-shared features, while the meta-adjuster is designed to learn domain-specific features. To enable this process, DomainMix is exploited to simulate data from diverse domains during teaching the meta-adjuster to adapt to the agnostic target domains. This learning mechanism urges the model to generalize to different agnostic target domains via adjusting the model without training. Extensive experiments demonstrate the effectiveness of our proposed method. Code is available: https://github.com/MetaVisionLab/DDG","doi":"10.24963/ijcai.2022/187","url":"https://ijcai.org/proceedings/2022/187","pdf":"https://ijcai.org/proceedings/2022/0187.pdf","markdown":"https://ijcai.org/proceedings/2022/187.md","bibtex":"https://ijcai.org/proceedings/2022/bibtex/187","volume":{"slug":"2022","title":"Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2022","url":"https://ijcai.org/proceedings/2022/","year":2022,"location":"Vienna","dates":"23-29 July 2022","editors":"Luc De Raedt","isbn":"978-1-956792-00-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2022"},"section":"Computer Vision","track":"Main Track","number":187,"pages":{"first":1342,"last":1348},"keywords":["Computer Vision: Transfer, low-shot, semi- and un- supervised learning"],"videos":[{"label":"Video #1","url":"https://slideslive.com/38986588","length_sec":90,"provider":"slideslive"},{"label":"Video #2","url":"https://slideslive.com/38984731","length_sec":368,"provider":"slideslive"}],"references":[{"position":1,"raw":"[Balaji et al., 2018] Yogesh Balaji, Swami Sankara- narayanan, and Rama Chellappa. Metareg: Towards do- main generalization using meta-regularization. NeurIPS, 2018.","title":"Metareg: Towards domain generalization using meta-regularization","authors":"Yogesh Balaji, Swami Sankaranarayanan, Rama Chellappa","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Carlucci et al., 2019] Fabio M Carlucci, Antonio D'Innocente, Silvia Bucci, Barbara Caputo, and Ta- tiana Tommasi. Domain generalization by solving jigsaw puzzles. In CVPR, 2019.","title":"Domain generalization by solving jigsaw puzzles","authors":"Fabio M Carlucci, D' Antonio, Silvia Innocente, Barbara Bucci, Tatiana Caputo, Tommasi","year":"2019","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Cha et al., 2021] Junbum Cha, Sanghyuk Chun, Kyungjae Lee, Han-Cheol Cho, Seunghyun Park, Yunsung Lee, and Sungrae Park. Swad: Domain generalization by seeking flat minima. NeurIPS, 2021.","title":"Swad: Domain generalization by seeking flat minima","authors":"Cha","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Chattopadhyay et al., 2020] Prithvijit Chattopadhyay, Yo- gesh Balaji, and Judy Hoffman. Learning to balance speci- ficity and invariance for in and out of domain generaliza- tion. In ECCV, 2020.","title":"Learning to balance specificity and invariance for in and out of domain generalization","authors":"Chattopadhyay","year":"2020","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Ding et al., 2019] Xiaohan Ding, Yuchen Guo, Guiguang Ding, and Jungong Han. Acnet: Strengthening the ker- nel skeletons for powerful cnn via asymmetric convolution blocks. In ICCV, 2019.","title":"Acnet: Strengthening the kernel skeletons for powerful cnn via asymmetric convolution blocks","authors":"Xiaohan Ding, Yuchen Guo, Guiguang Ding, Jungong Han","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Dou et al., 2019] Qi Dou, Daniel Coelho de Castro, Kon- stantinos Kamnitsas, and Ben Glocker. Domain general- ization via model-agnostic learning of semantic features. NeurIPS, 2019.","title":"Domain generalization via model-agnostic learning of semantic features","authors":"Dou","year":"2019","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[D'Innocente and Caputo, 2018] Antonio D'Innocente and Barbara Caputo. Domain generalization with domain- specific aggregation modules. In GCPR, 2018.","title":"Antonio D'Innocente and Barbara Caputo. Domain generalization with domainspecific aggregation modules","authors":"D'innocente And Caputo","year":"2018","venue":"GCPR","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Ganin and Lempitsky, 2015] Yaroslav Ganin and Victor Lempitsky. Unsupervised domain adaptation by backprop- agation. In ICML, 2015.","title":"Unsupervised domain adaptation by backpropagation","authors":"Yaroslav Ganin, Victor Lempitsky","year":"2015","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Ganin et al., 2016] Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, Franc ¸ois Laviolette, Mario Marchand, and Victor Lempitsky. Domain-adversarial training of neural networks. JMLR, 2016.","title":"Domain-adversarial training of neural networks","authors":"Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, Mario Franc ¸ois Laviolette, Victor Marchand, Lempitsky","year":"2016","venue":"JMLR","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Gulrajani and Lopez-Paz, 2021] Ishaan Gulrajani and David Lopez-Paz. In search of lost domain generalization. In ICLR, 2021.","title":"Ishaan Gulrajani and David Lopez-Paz. In search of lost domain generalization","authors":"Lopez-Paz Gulrajani","year":"2021","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Haeusser et al., 2017] Philip Haeusser, Thomas Frerix, Alexander Mordvintsev, and Daniel Cremers. Associative domain adaptation. In ICCV, 2017.","title":"Associative domain adaptation","authors":"Philip Haeusser, Thomas Frerix, Alexander Mordvintsev, Daniel Cremers","year":"2017","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[He et al., 2016] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recog- nition. In CVPR, 2016.","title":"Deep residual learning for image recognition","authors":"Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Hu et al., 2018] Jie Hu, Li Shen, and Gang Sun. Squeeze- and-excitation networks. In CVPR, 2018.","title":"Squeezeand-excitation networks","authors":"Jie Hu, Li Shen, Gang Sun","year":"2018","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Huang et al., 2020] Zeyi Huang, Haohan Wang, Eric P Xing, and Dong Huang. Self-challenging improves cross- domain generalization. In ECCV, 2020.","title":"Self-challenging improves crossdomain generalization","authors":"Zeyi Huang, Haohan Wang, Eric P Xing, Dong Huang","year":"2020","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Iwasawa and Matsuo, 2021] Yusuke Iwasawa and Yutaka Matsuo. Test-time classifier adjustment module for model- agnostic domain generalization. NeurIPS, 2021.","title":"Test-time classifier adjustment module for modelagnostic domain generalization","authors":"Matsuo Iwasawa, Yusuke Iwasawa, Yutaka Matsuo","year":"2021","venue":null,"doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Kim et al., 2021] Daehee Kim, Youngjun Yoo, Seunghyun Park, Jinkyu Kim, and Jaekoo Lee. Selfreg: Self- supervised contrastive regularization for domain general- ization. In ICCV, 2021.","title":"Selfreg: Selfsupervised contrastive regularization for domain generalization","authors":"Kim","year":"2021","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Li et al., 2017] Da Li, Yongxin Yang, Yi-Zhe Song, and Timothy M Hospedales. Deeper, broader and artier do- main generalization. In ICCV, 2017.","title":"Deeper, broader and artier domain generalization","authors":"Da Li, Yongxin Yang, Yi-Zhe Song, Timothy M Hospedales","year":"2017","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Li et al., 2018a] Da Li, Yongxin Yang, Yi-Zhe Song, and Timothy M Hospedales. Learning to generalize: Meta- learning for domain generalization. In AAAI, 2018.","title":"Learning to generalize: Metalearning for domain generalization","authors":"Li","year":"2018","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Li et al., 2018b] Ya Li, Xinmei Tian, Mingming Gong, Ya- jing Liu, Tongliang Liu, Kun Zhang, and Dacheng Tao. Deep domain generalization via conditional invariant ad- versarial networks. In ECCV, 2018.","title":"Deep domain generalization via conditional invariant adversarial networks","authors":"Li","year":"2018","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Li et al., 2019] Da Li, Jianshu Zhang, Yongxin Yang, Cong Liu, Yi-Zhe Song, and Timothy M Hospedales. Episodic training for domain generalization. In ICCV, 2019.","title":"Episodic training for domain generalization","authors":"Da Li, Jianshu Zhang, Yongxin Yang, Cong Liu, Yi-Zhe Song, Timothy M Hospedales","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Motiian et al., 2017] Saeid Motiian, Marco Piccirilli, Don- ald A Adjeroh, and Gianfranco Doretto. Unified deep su- pervised domain adaptation and generalization. In ICCV, 2017.","title":"Unified deep supervised domain adaptation and generalization","authors":"Saeid Motiian, Marco Piccirilli, Donald A Adjeroh, Gianfranco Doretto","year":"2017","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Nam et al., 2021] Hyeonseob Nam, HyunJae Lee, Jongchan Park, Wonjun Yoon, and Donggeun Yoo. Reducing do- main gap by reducing style bias. In CVPR, 2021.","title":"Reducing domain gap by reducing style bias","authors":"Nam","year":"2021","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Peng et al., 2019] Xingchao Peng, Qinxun Bai, Xide Xia, Zijun Huang, Kate Saenko, and Bo Wang. Moment match- ing for multi-source domain adaptation. In ICCV, 2019.","title":"Moment matching for multi-source domain adaptation","authors":"Peng","year":"2019","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Piratla et al., 2020] Vihari Piratla, Praneeth Netrapalli, and Sunita Sarawagi. Efficient domain generalization via common-specific low-rank decomposition. In ICML, 2020.","title":"Efficient domain generalization via common-specific low-rank decomposition","authors":"Praneeth Vihari Piratla, Sunita Netrapalli, Sarawagi","year":"2020","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Venkateswara et al., 2017] Hemanth Venkateswara, Jose Eusebio, Shayok Chakraborty, and Sethuraman Pan- chanathan. Deep hashing network for unsupervised do- main adaptation. In CVPR, 2017.","title":"Deep hashing network for unsupervised domain adaptation","authors":"Hemanth Venkateswara, Jose Eusebio, Shayok Chakraborty, Sethuraman Panchanathan","year":"2017","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Zhang et al., 2018] Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. mixup: Beyond empirical risk minimization. In ICLR, 2018.","title":"mixup: Beyond empirical risk minimization","authors":"Hongyi Zhang, Moustapha Cisse, David Yann N Dauphin, Lopez-Paz","year":"2018","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Zhou et al., 2020a] Kaiyang Zhou, Yongxin Yang, Timothy Hospedales, and Tao Xiang. Deep domain-adversarial im- age generation for domain generalisation. In AAAI, 2020.","title":"Deep domain-adversarial image generation for domain generalisation","authors":"Zhou","year":"2020","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Zhou et al., 2020b] Kaiyang Zhou, Yongxin Yang, Timothy Hospedales, and Tao Xiang. Learning to generate novel domains for domain generalization. In ECCV, 2020.","title":"Learning to generate novel domains for domain generalization","authors":"Zhou","year":"2020","venue":"ECCV","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Zhou et al., 2021] Kaiyang Zhou, Yongxin Yang, Yu Qiao, and Tao Xiang. Domain generalization with mixstyle. In ICLR, 2021.","title":"Domain generalization with mixstyle","authors":"Kaiyang Zhou, Yongxin Yang, Yu Qiao, Tao Xiang","year":"2021","venue":"ICLR","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}