On the estimation of entropy in the FastICA algorithm

JOURNAL OF MULTIVARIATE ANALYSIS

卷:181

Article

Issoglio, Elena¹  Smith, Paul¹  Voss, Jochen¹ 

[1] Univ Leeds, Sch Math, Leeds, W Yorkshire, England

关键词: Approximation;    Blind source separation;    Convergence;    Counterexample;    FastICA;    Independent component analysis;    Projection pursuit;    Projections;   

DOI  :  10.1016/j.jmva.2020.104689

来源: Elsevier

【 摘 要 】

The fastICA method is a popular dimension reduction technique used to reveal patterns in data. Here we show both theoretically and in practice that the approximations used in fastICA can result in patterns not being successfully recognised. We demonstrate this problem using a two-dimensional example where a clear structure is immediately visible to the naked eye, but where the projection chosen by fastICA fails to reveal this structure. This implies that care is needed when applying fastICA. We discuss how the problem arises and how it is intrinsically connected to the approximations that form the basis of the computational efficiency of fastICA. (C) 2020 The Authors. Published by Elsevier Inc.

【 授权许可】

Free