SIA OpenIR  > 机器人学研究室
Lifelong Metric Learning
Sun G(孙干)1,2; Cong Y(丛杨)1; Liu J(刘霁)3; Liu LQ(刘连庆)1; Xu XW(徐晓伟)4; Yu HB(于海斌)1
Department机器人学研究室
Source PublicationIEEE Transactions on Cybernetics
ISSN2168-2267
2019
Volume49Issue:8Pages:3168-3179
Indexed BySCI ; EI
EI Accession number20182705387212
WOS IDWOS:000467561700029
Contribution Rank1
Funding OrganizationNatural Science Foundation of China ; CAS-Youth Innovation Promotion Association Scholarship
KeywordLifelong Learning Metric Learning Multi-task Learning Low-rank Subspace
Abstract

The state-of-the-art online learning approaches are only capable of learning the metric for predefined tasks. In this paper, we consider a lifelong learning problem to mimic ``human learning,'' i.e., endowing a new capability to the learned metric for a new task from new online samples and incorporating the previous experiences. Therefore, we propose a new metric learning framework: lifelong metric learning (LML), which only utilizes the data of the new task to train the metric model while preserving the original capabilities. More specifically, the proposed LML maintains a common subspace for all learned metrics, named lifelong dictionary, transfers knowledge from the common subspace to learn each new metric learning task with task-specific idiosyncrasy, and redefines the common subspace over time to maximize performance across all metric tasks. For model optimization, we apply online passive aggressive optimization algorithm to achieve lifelong metric task learning, where the lifelong dictionary and task-specific partition are optimized alternatively and consecutively. Finally, we evaluate our approach by analyzing several multitask metric learning datasets. Extensive experimental results demonstrate effectiveness and efficiency of the proposed framework.

Language英语
WOS SubjectAutomation & Control Systems ; Computer Science, Artificial Intelligence ; Computer Science, Cybernetics
WOS Research AreaAutomation & Control Systems ; Computer Science
Funding ProjectNatural Science Foundation of China[61722311] ; Natural Science Foundation of China[U1613214] ; Natural Science Foundation of China[61533015] ; CAS-Youth Innovation Promotion Association Scholarship[2012163]
Citation statistics
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/22140
Collection机器人学研究室
Corresponding AuthorSun G(孙干)
Affiliation1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China
2.University of Chinese Academy of Sciences
3.Department of Computer Science, University of Rochester, Rochester, NY 14627 USA
4.Department of Information Science, University of Arkansas at Little Rock, Little Rock, AR 72204 USA
Recommended Citation
GB/T 7714
Sun G,Cong Y,Liu J,et al. Lifelong Metric Learning[J]. IEEE Transactions on Cybernetics,2019,49(8):3168-3179.
APA Sun G,Cong Y,Liu J,Liu LQ,Xu XW,&Yu HB.(2019).Lifelong Metric Learning.IEEE Transactions on Cybernetics,49(8),3168-3179.
MLA Sun G,et al."Lifelong Metric Learning".IEEE Transactions on Cybernetics 49.8(2019):3168-3179.
Files in This Item: Download All
File Name/Size DocType Version Access License
Lifelong Metric Lear(1858KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Sun G(孙干)]'s Articles
[Cong Y(丛杨)]'s Articles
[Liu J(刘霁)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Sun G(孙干)]'s Articles
[Cong Y(丛杨)]'s Articles
[Liu J(刘霁)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Sun G(孙干)]'s Articles
[Cong Y(丛杨)]'s Articles
[Liu J(刘霁)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Lifelong Metric Learning.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.