Generalization error bounds for kernel matrix completion and extrapolation
Machine Learning
2020-04-22 v1 Machine Learning
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.
Keywords
Cite
@article{arxiv.1906.08770,
title = {Generalization error bounds for kernel matrix completion and extrapolation},
author = {Pere Giménez-Febrer and Alba Pagès-Zamora and Georgios B. Giannakis},
journal= {arXiv preprint arXiv:1906.08770},
year = {2020}
}