Newton method for the ICA mixture model

Author

Palmer, J.A. ; Makeig, S. ; Kreutz-Delgado, K. ; Rao, B.D.

fYear

2008

fDate

March 31 2008-April 4 2008

Firstpage

1805

Lastpage

1808

Abstract

We derive an asymptotic Newton algorithm for quasi-maximum likelihood estimation of the ICA mixture model, using the ordinary gradient and Hessian. The probabilistic mixture framework yields an algorithm that can accommodate non-stationary environments and arbitrary source densities. We prove asymptotic stability when the source models match the true sources. An example application to EEC segmentation is given.

Keywords

Hessian matrices; Newton method; gradient methods; independent component analysis; maximum likelihood estimation; probability; signal processing; EEC segmentation; Hessian matrix; ICA mixture model; Newton method; asymptotic stability; gradient method; probabilistic mixture framework; quasimaximum likelihood estimation; signal processing; Asymptotic stability; Bayesian methods; Brain modeling; Electroencephalography; Independent component analysis; Large-scale systems; Newton method; Sensor arrays; Signal processing algorithms; Vectors; Bayesian linear mixture model; EEG signal analysis; Independent Component Analysis; Newton method;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2008. ICASSP 2008. IEEE International Conference on

Conference_Location

Las Vegas, NV

ISSN

1520-6149

Print_ISBN

978-1-4244-1483-3

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2008.4517982

Filename

4517982