Conference Paper

Convergence of the symmetrical FastICA algorithm

Neural Networks Res. Centre, Helsinki Univ. of Technol., Espoo, Finland
DOI: 10.1109/ICONIP.2002.1202844 Conference: Neural Information Processing, 2002. ICONIP '02. Proceedings of the 9th International Conference on, Volume: 3
Source: IEEE Xplore

ABSTRACT The FastICA algorithm is one of the most popular methods to solve problems in independent component analysis (ICA) and blind source separation. It has been shown experimentally that it outperforms most of the commonly used ICA algorithms in convergence speed. A rigorous convergence analysis has been presented only for the so-called one-unit case, in which just one of the rows of the separating matrix is considered. However, in the FastICA algorithm, there is also an explicit normalization step, and it may be questioned whether the extra rotation caused by the normalization will effect the convergence speed. The purpose of this paper is to show that this is not the case and the good convergence properties of the one-unit case are also shared by the full algorithm with symmetrical normalization.

  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: With the explosive growth of internet, web pages classification has become an essential issue. This is because web pages classification will provide an efficient information search to internet users. Without professional classification, a website would become a jumble yard of content which is confusing and time wasting. By using web pages classification, it allows web visitors to navigate a web site quickly and efficiently. However, presently most of the web directories are still being classified manually or using semi-automated (huge teams of human editors)[1]. Automated web pages classification is highly in demand in order to replace expensive manpower and reduce the time consumed. In this paper we analyze the concept of a new model, which uses an integration of Principal Component Analysis (PCA) and Independent Component Analysis (ICA) as feature reduction for web pages classification. This model consists of several modules, which are web page retrieval process, stemming, stop-word filtering, feature reduction, feature selection, classification and evaluation.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The information theoretic concept of mutual information provides a general framework to evaluate dependencies between variables. Its estimation however using B-Spline has not been used before in creating an approach for Independent Component Analysis. In this paper we present a B-Spline estimator for mutual information to find the independent components in mixed signals. Tested using electroencephalography (EEG) signals the resulting BMICA (B-Spline Mutual Information Independent Component Analysis) exhibits better performance than the standard Independent Component Analysis algorithms of FastICA, JADE, SOBI and EFICA in similar simulations. BMICA was found to be also more reliable than the "renown" FastICA. .
    Signal & Image Processing : An International Journal. 04/2012; 3(2).