A novel dimension reduction procedure for searching non-gaussian subspaces

Authors:
Motoaki Kawanabe;Gilles Blanchard;Masashi Sugiyama;Vladimir Spokoiny;Klaus-Robert Müller
Affiliations:
Fraunhofer FIRST.IDA, Germany;Fraunhofer FIRST.IDA, Germany;Department of Computer Science, Tokyo Institute of Technology, Japan;Weierstrass Institute and Humboldt University, Germany;Fraunhofer FIRST.IDA, Germany
Venue:
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Year:
2006

Citing 7
Cited 0

Independent component analysis, a new concept?

Signal Processing - Special issue on higher order statistics
GTM: the generative topographic mapping

Neural Computation
Kernel-based nonlinear blind source separation

Neural Computation
Laplacian Eigenmaps for dimensionality reduction and data representation

Neural Computation
Estimating Functions for Blind Separation When Sources Have Variance Dependencies

The Journal of Machine Learning Research
A Projection Pursuit Algorithm for Exploratory Data Analysis

IEEE Transactions on Computers
Fast and robust fixed-point algorithms for independent component analysis

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this article, we consider high-dimensional data which contains a low-dimensional non-Gaussian structure contaminated with Gaussian noise and propose a new linear method to identify the non-Gaussian subspace. Our method NGCA (Non-Gaussian Component Analysis) is based on a very general semi-parametric framework and has a theoretical guarantee that the estimation error of finding the non-Gaussian components tends to zero at a parametric rate. NGCA can be used not only as preprocessing for ICA, but also for extracting and visualizing more general structures like clusters. A numerical study demonstrates the usefulness of our method.