Orthogonal Extended Infomax Algorithm

TL;DR

This paper introduces an orthogonal extended infomax ICA algorithm that converges faster than traditional methods by using a multiplicative orthogonal-group update scheme, improving performance on EEG data analysis.

Contribution

The paper presents a novel orthogonal extended infomax ICA algorithm (OgExtInf) that accelerates convergence using an orthogonal-group based update scheme, outperforming existing ICA algorithms.

Findings

OgExtInf converges faster than traditional extended infomax.

OgExtInf outperforms FastICA and Picard on EEG datasets.

Improved ICA method benefits online EEG processing applications.

Abstract

The extended infomax algorithm for independent component analysis (ICA) can separate sub- and super-Gaussian signals but converges slowly as it uses stochastic gradient optimization. In this paper, an improved extended infomax algorithm is presented that converges much faster. Accelerated convergence is achieved by replacing the natural gradient learning rule of extended infomax by a fully-multiplicative orthogonal-group based update scheme of the unmixing matrix leading to an orthogonal extended infomax algorithm (OgExtInf). Computational performance of OgExtInf is compared with two fast ICA algorithms: the popular FastICA and Picard, a L-BFGS algorithm belonging to the family of quasi-Newton methods. Our results demonstrate superior performance of the proposed method on small-size EEG data sets as used for example in online EEG processing systems, such as brain-computer interfaces or clinical systems for spike and seizure detection.