Name
Abstract | ||
|---|---|---|
Deflation-based FastICA, where independent components (IC's) are extracted one-by-one, is among the most popular methods for estimating an unmixing matrix in the independent component analysis (ICA) model. In the literature, it is often seen rather as an algorithm than an estimator related to a certain objective function, and only recently has its statistical properties been derived. One of the recent findings is that the order, in which the independent components are extracted in practice, has a strong effect on the performance of the estimator. In this paper we review these recent findings and propose a new “reloaded” procedure to ensure that the independent components are extracted in an optimal order. The reloaded algorithm improves the separation performance of the deflation-based FastICA estimator as amply illustrated by our simulation studies. Reloading also seems to render the algorithm more stable. |
Year | Venue | Field |
|---|---|---|
2011 | European Signal Processing Conference | Matrix (mathematics),Algorithm,Robustness (computer science),Deflation,Independent component analysis,FastICA,Artificial intelligence,Limiting,Machine learning,Mathematics,Estimator |
DocType | ISSN | Citations |
Conference | 2076-1465 | 6 |
PageRank | References | Authors |
0.66 | 5 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Klaus Nordhausen | 1 | 90 | 14.33 |
| Pauliina Ilmonen | 2 | 17 | 2.63 |
| Abhijit Mandal | 3 | 6 | 0.66 |
| Hannu Oja | 4 | 88 | 13.07 |
| Esa Ollila | 5 | 351 | 33.51 |