Loading...
Loading...

Go to the content (press return)

Generalization error bounds for kernel matrix completion and extrapolation

Author
Gimenez, P.; Pagès-Zamora, A.; Giannakis, G.B.
Type of activity
Journal article
Journal
IEEE signal processing letters
Date of publication
2020-01-27
Volume
27
Number
1
First page
326
Last page
330
DOI
10.1109/LSP.2020.2970306
Project funding
Coding and Signal Processing for Emerging Wireless Communication and Sensor Networks
Distributed techniques for the management and operation of wireless cellular networks, sensor networks and the smart energy grid
Grup de processament del senyal i comunicacions
Repository
http://hdl.handle.net/2117/191729 Open in new window
https://arxiv.org/abs/1906.08770 Open in new window
URL
https://ieeexplore.ieee.org/abstract/document/8974415 Open in new window
Abstract
Prior information can be incorporated in matrix completion to improve estimation accuracy and extrapolate the missing entries. Reproducing kernel Hilbert spaces provide tools to leverage the said prior information, and derive more reliable algorithms. This paper analyzes the generalization error of such approaches, and presents numerical tests confirming the theoretical results © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in ...
Citation
Gimenez, P.; Pagès-Zamora, A.; Giannakis, G.B. Generalization error bounds for kernel matrix completion and extrapolation. "IEEE signal processing letters", 27 Gener 2020, vol. 27, núm. 1, p. 326-330.
Keywords
Generalization error, Kernel regression, Matrix completion, Rademacher complexity
Group of research
SPCOM - Signal Processing and Communications Group

Participants

Attachments