A Bias Trick for Centered Robust Principal Component Analysis

11/19/2019

∙

by   Baokun He, et al.

∙

0

∙

share

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.