Independent Component Analysis based on multiple data-weighting

TL;DR

This paper introduces MWeICA, a novel ICA algorithm that uses multiple data-weighting and approximate covariance matrix diagonalization, improving independence extraction with comparable computational efficiency.

Contribution

The paper presents a new ICA method based on weighted covariance matrices and theoretical guarantees of independence, outperforming existing ICA techniques.

Findings

MWeICA achieves better independence separation than state-of-the-art methods.

MWeICA maintains similar computational time to existing ICA algorithms.

Experimental results validate the effectiveness of the proposed approach.

Abstract

Independent Component Analysis (ICA) - one of the basic tools in data analysis - aims to find a coordinate system in which the components of the data are independent. In this paper we present Multiple-weighted Independent Component Analysis (MWeICA) algorithm, a new ICA method which is based on approximate diagonalization of weighted covariance matrices. Our idea is based on theoretical result, which says that linear independence of weighted data (for gaussian weights) guarantees independence. Experiments show that MWeICA achieves better results to most state-of-the-art ICA methods, with similar computational time.