WhiteningFree LeastSquares NonGaussian Component Analysis
[edit]
Proceedings of the Ninth Asian Conference on Machine Learning, PMLR 77:375390, 2017.
Abstract
\emphNonGaussian component analysis (NGCA) is an unsupervised linear dimension reduction method that extracts lowdimensional nonGaussian “signals” from highdimensional data contaminated with Gaussian noise. NGCA can be regarded as a generalization of \emphprojection pursuit (PP) and \emphindependent component analysis (ICA) to multidimensional and dependent nonGaussian components. Indeed, seminal approaches to NGCA are based on PP and ICA. Recently, a novel NGCA approach called \emphleastsquares NGCA (LSNGCA) has been developed, which gives a solution analytically through leastsquares estimation of \emphlogdensity gradients and eigendecomposition. However, since \emphprewhitening of data is involved in LSNGCA, it performs unreliably when the data covariance matrix is illconditioned, which is often the case in highdimensional data analysis. In this paper, we propose a \emphwhiteningfree variant of LSNGCA and experimentally demonstrate its superiority.
Related Material


