Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators

All the algorithms for ICA require high-order statistics to estimate the independent components. This is because second-order information is insufficient to assess that two random variables are independent of each other. It is known that the robustness of the high-order sample estimators is poor, me...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Messina, Francisco, Cernuschi-Frías, Bruno
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2012
Materias:
Ciencias Informáticas
ICA
Algorithm
Batch and Adaptive MMSE Estimators
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/123933
https://41jaiio.sadio.org.ar/sites/default/files/21_AST_2012.pdf
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-123933
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
ICA
Algorithm
Batch and Adaptive MMSE Estimators
spellingShingle Ciencias Informáticas
ICA
Algorithm
Batch and Adaptive MMSE Estimators
Messina, Francisco
Cernuschi-Frías, Bruno
Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators
topic_facet Ciencias Informáticas
ICA
Algorithm
Batch and Adaptive MMSE Estimators
description All the algorithms for ICA require high-order statistics to estimate the independent components. This is because second-order information is insufficient to assess that two random variables are independent of each other. It is known that the robustness of the high-order sample estimators is poor, meaning that a few outliers can change dramatically its value. In this paper, we generalize the alternative robust statistics for moments and cumulants introduced by Welling presenting the MMSE-robust moments. Then we present a batch and adaptive versions of an algorithm for estimating the parameters that define the estimator. Finally, we modify two FastICA algorithms of ICA based on kurtosis and negentropy to apply the MMSE robust estimators and show some experiments with supergaussian sources to demonstrate the improvement.
format Objeto de conferencia
Objeto de conferencia
author Messina, Francisco
Cernuschi-Frías, Bruno
author_facet Messina, Francisco
Cernuschi-Frías, Bruno
author_sort Messina, Francisco
title Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators
title_short Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators
title_full Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators
title_fullStr Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators
title_full_unstemmed Robust Parallel Fast-ICA Algorithms Using Batch and Adaptive MMSE Estimators
title_sort robust parallel fast-ica algorithms using batch and adaptive mmse estimators
publishDate 2012
url http://sedici.unlp.edu.ar/handle/10915/123933
https://41jaiio.sadio.org.ar/sites/default/files/21_AST_2012.pdf
work_keys_str_mv AT messinafrancisco robustparallelfasticaalgorithmsusingbatchandadaptivemmseestimators
AT cernuschifriasbruno robustparallelfasticaalgorithmsusingbatchandadaptivemmseestimators
bdutipo_str Repositorios
_version_ 1764820450537373698

Ejemplares similares