Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/91840
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Department of Applied Mathematics | en_US |
| dc.creator | Chen, X | en_US |
| dc.creator | Tang, B | en_US |
| dc.creator | Fan, J | en_US |
| dc.creator | Guo, X | en_US |
| dc.date.accessioned | 2021-12-23T02:14:45Z | - |
| dc.date.available | 2021-12-23T02:14:45Z | - |
| dc.identifier.issn | 0885-064X | en_US |
| dc.identifier.uri | http://hdl.handle.net/10397/91840 | - |
| dc.language.iso | en | en_US |
| dc.publisher | Academic Press | en_US |
| dc.subject | Learning theory | en_US |
| dc.subject | Online learning | en_US |
| dc.subject | Gradient descent | en_US |
| dc.subject | Reproducing kernel Hilbert space | en_US |
| dc.subject | Error analysis | en_US |
| dc.title | Online gradient descent algorithms for functional data learning | en_US |
| dc.type | Journal/Magazine Article | en_US |
| dc.identifier.doi | 10.1016/j.jco.2021.101635 | en_US |
| dcterms.abstract | Functional linear model is a fruitfully applied general framework for regression problems, including those with intrinsically infinite-dimensional data. Online gradient descent methods, despite their evidenced power of processing online or large-sized data, are not well studied for learning with functional data. In this paper, we study reproducing kernel-based online learning algorithms for functional data, and derive convergence rates for the expected excess prediction risk under both online and finite-horizon settings of step-sizes respectively. It is well understood that nontrivial uniform convergence rates for the estimation task depend on the regularity of the slope function. Surprisingly, the convergence rates we derive for the prediction task can assume no regularity from slope. Our analysis reveals the intrinsic difference between the estimation task and the prediction task in functional data learning. | en_US |
| dcterms.accessRights | embargoed access | en_US |
| dcterms.bibliographicCitation | Journal of complexity, 2021, In Press, 101635, https://doi.org/10.1016/j.jco.2021.101635 | en_US |
| dcterms.isPartOf | Journal of complexity | en_US |
| dcterms.issued | 2021 | - |
| dc.identifier.eissn | 1090-2708 | en_US |
| dc.identifier.artn | 101635 | en_US |
| dc.description.validate | 202112 bcvc | en_US |
| dc.description.oa | Not applicable | en_US |
| dc.identifier.FolderNumber | a1122-n01 | - |
| dc.identifier.SubFormID | 43965 | - |
| dc.description.fundingSource | RGC | en_US |
| dc.description.fundingSource | Others | en_US |
| dc.description.fundingText | RGC: 15304917 | en_US |
| dc.description.fundingText | Others: ZE8Q | en_US |
| dc.description.pubStatus | Early release | en_US |
| dc.date.embargo | 0000-00-00 (to be updated) | en_US |
| Appears in Collections: | Journal/Magazine Article | |
Access
View full-text via PolyU eLinks 
Page views
86
Last Week
1
1
Last month
Citations as of Oct 1, 2023
SCOPUSTM
Citations
5
Citations as of Sep 28, 2023
WEB OF SCIENCETM
Citations
3
Citations as of Sep 28, 2023
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.