SIA OpenIR  > 机器人学研究室
Multilinear Multitask Learning by Rank-Product Regularization
Zhao Q(赵谦)1; Rui, Xiangyu1; Han Z(韩志)2; Meng DY(孟德宇)1,2,3
Department机器人学研究室
Source PublicationIEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
ISSN2162-237X
2020
Volume31Issue:4Pages:1336-1350
Indexed BySCI ; EI
EI Accession number20201608414415
WOS IDWOS:000525351800022
Contribution Rank2
Funding OrganizationChina NSFC ProjectNational Natural Science Foundation of China [61603292, 61661166011, 11690011, 61721002, 61773367, U1811461] ; Youth Innovation Promotion Association of the Chinese Academy of Sciences [2016183] ; State Key Laboratory of Robotics [2017-O09]
KeywordMultilinear multitask learning (MTL) rank product tensor sparsity
Abstract

Multilinear multitask learning (MLMTL) considers an MTL problem in which tasks are arranged by multiple indices. By exploiting the higher order correlations among the tasks, MLMTL is expected to improve the performance of traditional MTL, which only considers the first-order correlation across all tasks, e.g., low-rank structure of the coefficient matrix. The key to MLMTL is designing a rational regularization term to represent the latent correlation structure underlying the coefficient tensor instead of matrix. In this paper, we propose a new MLMTL model by employing the rank-product regularization term in the objective, which on one hand can automatically rectify the weights along all its tensor modes and on the other hand have an explicit physical meaning. By using this regularization, the intrinsic high-order correlations among tasks can be more precisely described, and thus, the overall performance of all tasks can be improved. To solve the resulted optimization model, we design an efficient algorithm by applying the alternating direction method of multipliers (ADMM). We also analyze the convergence and show that the proposed algorithm, with certain restriction, is asymptotically regular. Experiments on both synthetic and real data sets substantiate the superiority of the proposed method beyond the existing MLMTL methods in terms of accuracy and efficiency.

Language英语
WOS SubjectComputer Science, Artificial Intelligence ; Computer Science, Hardware & Architecture ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS Research AreaComputer Science ; Engineering
Funding ProjectChina NSFC Project[61603292] ; China NSFC Project[61661166011] ; China NSFC Project[11690011] ; China NSFC Project[61721002] ; China NSFC Project[61773367] ; China NSFC Project[U1811461] ; Youth Innovation Promotion Association of the Chinese Academy of Sciences[2016183] ; State Key Laboratory of Robotics[2017-O09]
Citation statistics
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/26724
Collection机器人学研究室
Corresponding AuthorMeng DY(孟德宇)
Affiliation1.School of Mechanical Engineering, Shenyang Jianzhu University, Shenyang 110168, China
2.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China
3.Department of Neurology, People's Hospital of Liaoning Province, Shenyang 110016, China
4.School of Electromechanical and Automotive Engineering, Yantai University, Yantai 264005, China
Recommended Citation
GB/T 7714
Zhao Q,Rui, Xiangyu,Han Z,et al. Multilinear Multitask Learning by Rank-Product Regularization[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2020,31(4):1336-1350.
APA Zhao Q,Rui, Xiangyu,Han Z,&Meng DY.(2020).Multilinear Multitask Learning by Rank-Product Regularization.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,31(4),1336-1350.
MLA Zhao Q,et al."Multilinear Multitask Learning by Rank-Product Regularization".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 31.4(2020):1336-1350.
Files in This Item:
File Name/Size DocType Version Access License
Multilinear Multitas(2338KB)期刊论文出版稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhao Q(赵谦)]'s Articles
[Rui, Xiangyu]'s Articles
[Han Z(韩志)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhao Q(赵谦)]'s Articles
[Rui, Xiangyu]'s Articles
[Han Z(韩志)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhao Q(赵谦)]'s Articles
[Rui, Xiangyu]'s Articles
[Han Z(韩志)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Multilinear Multitask Learning by Rank-Product Regularization.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

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