Skip to yearly menu bar Skip to main content


Correlated-PCA: Principal Components' Analysis when Data and Noise are Correlated

Namrata Vaswani · Han Guo

Area 5+6+7+8 #162

Keywords: [ Component Analysis (ICA,PCA,CCA, FLDA) ] [ Large Scale Learning and Big Data ]


Given a matrix of observed data, Principal Components Analysis (PCA) computes a small number of orthogonal directions that contain most of its variability. Provably accurate solutions for PCA have been in use for a long time. However, to the best of our knowledge, all existing theoretical guarantees for it assume that the data and the corrupting noise are mutually independent, or at least uncorrelated. This is valid in practice often, but not always. In this paper, we study the PCA problem in the setting where the data and noise can be correlated. Such noise is often also referred to as ``data-dependent noise". We obtain a correctness result for the standard eigenvalue decomposition (EVD) based solution to PCA under simple assumptions on the data-noise correlation. We also develop and analyze a generalization of EVD, cluster-EVD, that improves upon EVD in certain regimes.

Live content is unavailable. Log in and register to view live content