Kernel robust singular value decomposition

作者：

Highlights：

• Four kernel robust singular value decomposition (KR-SVD) algorithms are proposed.

• The KR-SVD allows to obtain robust estimates for the singular values and vectors.

• The KR-SVD outperforms the standard SVD and other robust SVD algorithms.

• The KR-SVD performs well in image recovery when contaminated with noise.

摘要

•Four kernel robust singular value decomposition (KR-SVD) algorithms are proposed.•The KR-SVD allows to obtain robust estimates for the singular values and vectors.•The KR-SVD outperforms the standard SVD and other robust SVD algorithms.•The KR-SVD performs well in image recovery when contaminated with noise.

论文关键词：Singular value decomposition,Kernel functions,Outlier,Robust regression,Robust SVD

论文评审过程：Received 17 March 2022, Revised 21 July 2022, Accepted 12 August 2022, Available online 18 August 2022, Version of Record 26 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118555