Towards Corruption-Agnostic Robust Domain Adaptation
文献类型:期刊论文
作者 | Xu, Yifan1,5; Sheng, Kekai4; Dong, Weiming1,3; Wu, Baoyuan2; Xu, Changsheng1,5; Hu, Bao-Gang1 |
刊名 | ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS |
出版日期 | 2022-11-01 |
卷号 | 18期号:4页码:16 |
ISSN号 | 1551-6857 |
关键词 | Domain adaptation corruption robustness transfer learning |
DOI | 10.1145/3501800 |
英文摘要 | Great progress has been achieved in domain adaptation in decades. Existing works are always based on an ideal assumption that testing target domains are independent and identically distributed with training target domains. However, due to unpredictable corruptions (e.g., noise and blur) in real data, such as web images and real-world object detection, domain adaptation methods are increasingly required to be corruption robust on target domains. We investigate a new task, corruption-agnostic robust domain adaptation (CRDA), to be accurate on original data and robust against unavailable-for-training corruptions on target domains. This task is non-trivial due to the large domain discrepancy and unsupervised target domains. We observe that simple combinations of popular methods of domain adaptation and corruption robustness have suboptimal CRDA results. We propose a newapproach based on two technical insights into CRDA, as follows: (1) an easy-to-plug module called domain discrepancy generator (DDG) that generates samples that enlarge domain discrepancy to mimic unpredictable corruptions; (2) a simple but effective teacher-student scheme with contrastive loss to enhance the constraints on target domains. Experiments verify that DDG maintains or even improves its performance on original data and achieves better corruption robustness than baselines. Our code is available at: https://github.com/YifanXu74/CRDA. |
资助项目 | National Natural Science Foundation of China[U20B2070] ; National Natural Science Foundation of China[61832016] ; National Natural Science Foundation of China[62036012] ; National Natural Science Foundation of China[61720106006] ; CASIA-Tencent Youtu joint research project |
WOS研究方向 | Computer Science |
语种 | 英语 |
出版者 | ASSOC COMPUTING MACHINERY |
WOS记录号 | WOS:000776441600010 |
资助机构 | National Natural Science Foundation of China ; CASIA-Tencent Youtu joint research project |
源URL | [http://ir.ia.ac.cn/handle/173211/48204] |
专题 | 自动化研究所_模式识别国家重点实验室_多媒体计算与图形学团队 |
通讯作者 | Dong, Weiming |
作者单位 | 1.Chinese Acad Sci, NLPR, Inst Automat, 95 East Zhongguancun Rd, Beijing 100190, Peoples R China 2.Chinese Univ Hong Kong, Shenzhen Res Inst Big Data, 2001 Longxiang Rd, Shenzhen 518172, Peoples R China 3.CASIA LLvis Joint Lab, 95 East Zhongguancun Rd, Beijing 100190, Peoples R China 4.Tencent Inc, Youtu Lab, 397 Tianlin Rd, Shanghai 201103, Peoples R China 5.Univ Chinese Acad Sci, Sch Artificial Intelligence, 95 East Zhongguancun Rd, Beijing 100190, Peoples R China |
推荐引用方式 GB/T 7714 | Xu, Yifan,Sheng, Kekai,Dong, Weiming,et al. Towards Corruption-Agnostic Robust Domain Adaptation[J]. ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS,2022,18(4):16. |
APA | Xu, Yifan,Sheng, Kekai,Dong, Weiming,Wu, Baoyuan,Xu, Changsheng,&Hu, Bao-Gang.(2022).Towards Corruption-Agnostic Robust Domain Adaptation.ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS,18(4),16. |
MLA | Xu, Yifan,et al."Towards Corruption-Agnostic Robust Domain Adaptation".ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS 18.4(2022):16. |
入库方式: OAI收割
来源:自动化研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。