Evolving Metric Learning for Incremental and Decremental Features | |
Dong JH(董家华)1,2,3; Cong Y(丛杨)2,3; Sun G(孙干)2,3; Zhang T(张涛)1,2,3; Tang X(唐旭)2,3; Xu XW(徐晓伟)4 | |
刊名 | IEEE Transactions on Circuits and Systems for Video Technology |
2022 | |
卷号 | 32期号:4页码:2290-2302 |
关键词 | Data models Extraterrestrial measurements Feature extraction instance and feature evolutions low-rank constraint Measurement Online metric learning Optimization Robot sensing systems smoothed Wasserstein distance Task analysis |
ISSN号 | 1051-8215 |
产权排序 | 1 |
英文摘要 | Online metric learning has been widely exploited for large-scale data classification due to the low computational cost. However, amongst online practical scenarios where the features are evolving (e.g., some features are vanished and some new features are augmented), most metric learning models cannot be successfully applied to these scenarios, although they can tackle the evolving instances efficiently. To address the challenge, we develop a new online Evolving Metric Learning (EML) model for incremental and decremental features, which can handle the instance and feature evolutions simultaneously by incorporating with a smoothed Wasserstein metric distance. Specifically, our model contains two essential stages: a Transforming stage (T-stage) and a Inheriting stage (I-stage). For the T-stage, we propose to extract important information from vanished features while neglecting non-informative knowledge, and forward it into survived features by transforming them into a low-rank discriminative metric space. It further explores the intrinsic low-rank structure of heterogeneous samples to reduce the computation and memory burden especially for highly-dimensional large-scale data. For the I-stage, we inherit the metric performance of survived features from the T-stage and then expand to include the new augmented features. Moreover, a smoothed Wasserstein distance is utilized to characterize the similarity relationships among the heterogeneous and complex samples, since the evolving features are not strictly aligned in the different stages. In addition to tackling the challenges in one-shot case, we also extend our model into multi-shot scenario. After deriving an efficient optimization strategy for both T-stage and I-stage, extensive experiments on several datasets verify the superior performance of our EML model. IEEE |
资助项目 | National Key Research and Development Program of China[2019YFB1310300] ; National Nature Science Foundation of China[61722311] ; National Nature Science Foundation of China[61821005] ; National Nature Science Foundation of China[62003336] ; Nature Foundation of Liaoning Province of China[2020-KF-11-01] |
WOS关键词 | SIMILARITY |
WOS研究方向 | Engineering |
语种 | 英语 |
WOS记录号 | WOS:000778973700048 |
资助机构 | National Key Research and Development Program of China (2019YFB1310300) ; National Nature Science Foundation of China under Grant (61722311, 61821005, 62003336) |
内容类型 | 期刊论文 |
源URL | [http://ir.sia.cn/handle/173321/29503] |
专题 | 沈阳自动化研究所_机器人学研究室 |
通讯作者 | Cong Y(丛杨) |
作者单位 | 1.University of Chinese Academy of Sciences, Beijing 100049, China 2.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, China 3.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China 4.Department of Information Science, University of Arkansas at Little Rock, Arkansas 72204, USA. |
推荐引用方式 GB/T 7714 | Dong JH,Cong Y,Sun G,et al. Evolving Metric Learning for Incremental and Decremental Features[J]. IEEE Transactions on Circuits and Systems for Video Technology,2022,32(4):2290-2302. |
APA | Dong JH,Cong Y,Sun G,Zhang T,Tang X,&Xu XW.(2022).Evolving Metric Learning for Incremental and Decremental Features.IEEE Transactions on Circuits and Systems for Video Technology,32(4),2290-2302. |
MLA | Dong JH,et al."Evolving Metric Learning for Incremental and Decremental Features".IEEE Transactions on Circuits and Systems for Video Technology 32.4(2022):2290-2302. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论