SIA OpenIR  > 机器人学研究室
Learning Through Deterministic Assignment of Hidden Parameters
Fang, Jian1; Lin SB(林绍波)2,3; Xu ZB(徐宗本)1
Department机器人学研究室
Source PublicationIEEE Transactions on Cybernetics
ISSN2168-2267
2018
Pages1-14
Indexed ByEI
EI Accession number20185206300246
Contribution Rank2
Funding OrganizationNational Natural Science Foundation of China under Grant 61876133 and Grant 11771021 ; State Key Laboratory of Robotics (2018-05)
KeywordBright parameters hidden parameters learning rate neural networks supervised learning
AbstractSupervised learning frequently boils down to determining hidden and bright parameters in a parameterized hypothesis space based on finite input-output samples. The hidden parameters determine the nonlinear mechanism of an estimator, while the bright parameters characterize the linear mechanism. In a traditional learning paradigm, hidden and bright parameters are not distinguished and trained simultaneously in one learning process. Such a one-stage learning (OSL) brings a benefit of theoretical analysis but suffers from the high computational burden. In this paper, we propose a two-stage learning scheme, learning through deterministic assignment of hidden parameters (LtDaHPs), suggesting to deterministically generate the hidden parameters by using minimal Riesz energy points on a sphere and equally spaced points in an interval. We theoretically show that with such a deterministic assignment of hidden parameters, LtDaHP with a neural network realization almost shares the same generalization performance with that of OSL. Then, LtDaHP provides an effective way to overcome the high computational burden of OSL. We present a series of simulations and application examples to support the outperformance of LtDaHP.
Language英语
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/23937
Collection机器人学研究室
Corresponding AuthorLin SB(林绍波)
Affiliation1.School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an 710048, China.
2.Department of Mathematics, Wenzhou University, Wenzhou 325035, China
3.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China (e-mail: sblin1983@gmail.com).
Recommended Citation
GB/T 7714
Fang, Jian,Lin SB,Xu ZB. Learning Through Deterministic Assignment of Hidden Parameters[J]. IEEE Transactions on Cybernetics,2018:1-14.
APA Fang, Jian,Lin SB,&Xu ZB.(2018).Learning Through Deterministic Assignment of Hidden Parameters.IEEE Transactions on Cybernetics,1-14.
MLA Fang, Jian,et al."Learning Through Deterministic Assignment of Hidden Parameters".IEEE Transactions on Cybernetics (2018):1-14.
Files in This Item: Download All
File Name/Size DocType Version Access License
期刊论文出版稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Fang, Jian]'s Articles
[Lin SB(林绍波)]'s Articles
[Xu ZB(徐宗本)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Fang, Jian]'s Articles
[Lin SB(林绍波)]'s Articles
[Xu ZB(徐宗本)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Fang, Jian]'s Articles
[Lin SB(林绍波)]'s Articles
[Xu ZB(徐宗本)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Learning Through Deterministic Assignment of Hidden Parameters.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

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