Paper
11 June 2015 Auxiliary function approach to independent component analysis and independent vector analysis
Author Affiliations +
Abstract
In this paper, we review an auxiliary function approach to independent component analysis (ICA) and independent vector analysis (IVA). The derived algorithm consists of two alternative updates: 1) weighted covariance matrix update and 2) demixing matrix update, which include no tuning parameters such as a step size in the gradient descent method. The monotonic decrease of the objective function is guaranteed by the principle of the auxiliary function method. The experimental evaluation shows that the derived update rules yield faster convergence and better results than natural gradient updates. An efficient implementation on a mobile phone is also presented.
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
N. Ono "Auxiliary function approach to independent component analysis and independent vector analysis", Proc. SPIE 9496, Independent Component Analyses, Compressive Sampling, Large Data Analyses (LDA), Neural Networks, Biosystems, and Nanoengineering XIII, 94960L (11 June 2015); https://doi.org/10.1117/12.2179859
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Independent component analysis

Algorithm development

Cell phones

Stochastic processes

Data analysis

Information science

Matrices

Back to Top