中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 中科院工业视觉智能装备工程实验室
Bridging the gap with grad: Integrating active learning into semi-supervised domain generalization

文献类型：期刊论文

作者Li, Jingwei2,3; Li, Yuan2,3; Tan, Jie3; Liu, Chengbao1,2,3
刊名NEURAL NETWORKS
出版日期2024-03-01
卷号171页码:186-199
ISSN号0893-6080
关键词Domain generalization Semi-supervised learning Active learning
DOI10.1016/j.neunet.2023.12.017
通讯作者Liu, Chengbao(liuchengbao2016@163.com)
英文摘要Domain generalization (DG) aims to generalize from a large amount of source data that are fully annotated. However, it is laborious to collect labels for all source data in practice. Some research gets inspiration from semi-supervised learning (SSL) and develops a new task called semi-supervised domain generalization (SSDG). Unlabeled source data is trained jointly with labeled one to significantly improve the performance. Nevertheless, different research adopts different settings, leading to unfair comparisons. Moreover, the initial annotation of unlabeled source data is random, causing unstable and unreliable training. To this end, we first specify the training paradigm, and then leverage active learning (AL) to handle the issues. We further develop a new task called Active Semi-supervised Domain Generalization (ASSDG), which consists of two parts, i.e., SSDG and AL. We delve deep into the commonalities of SSL and AL and propose a unified framework called Gradient-Similarity-based Sample Filtering and Sorting (GSSFS) to iteratively train the SSDG and AL parts. Gradient similarity is utilized to select reliable and informative unlabeled source samples for these two parts respectively. Our methods are simple yet efficient, and extensive experiments demonstrate that our methods can achieve the best results on the DG datasets in the low-data regime without bells and whistles.
资助项目National Nature Science Foundation of China[62003344] ; National Key Research and Development Program of China[2022YFB 3304602]
WOS研究方向Computer Science ; Neurosciences & Neurology
语种英语
出版者PERGAMON-ELSEVIER SCIENCE LTD
WOS记录号WOS:001140720500001
资助机构National Nature Science Foundation of China ; National Key Research and Development Program of China
源URL[http://ir.ia.ac.cn/handle/173211/55422]  
专题中科院工业视觉智能装备工程实验室
通讯作者Liu, Chengbao
作者单位1.Chinese Acad Sci, Inst Automat, 95 East Zhongguancun Rd, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
3.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714		 Li, Jingwei,Li, Yuan,Tan, Jie,et al. Bridging the gap with grad: Integrating active learning into semi-supervised domain generalization[J]. NEURAL NETWORKS,2024,171:186-199.
APA Li, Jingwei,Li, Yuan,Tan, Jie,&Liu, Chengbao.(2024).Bridging the gap with grad: Integrating active learning into semi-supervised domain generalization.NEURAL NETWORKS,171,186-199.
MLA Li, Jingwei,et al."Bridging the gap with grad: Integrating active learning into semi-supervised domain generalization".NEURAL NETWORKS 171(2024):186-199.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
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。