English

A Bias Trick for Centered Robust Principal Component Analysis

Machine Learning 2019-11-20 v1 Machine Learning

Abstract

Outlier based Robust Principal Component Analysis (RPCA) requires centering of the non-outliers. We show a "bias trick" that automatically centers these non-outliers. Using this bias trick we obtain the first RPCA algorithm that is optimal with respect to centering.

Keywords

Cite

@article{arxiv.1911.08024,
  title  = {A Bias Trick for Centered Robust Principal Component Analysis},
  author = {Baokun He and Guihong Wan and Haim Schweitzer},
  journal= {arXiv preprint arXiv:1911.08024},
  year   = {2019}
}
R2 v1 2026-06-23T12:20:07.104Z