Title :
Two approaches to estimation of overcomplete independent component bases
Author :
Inki, Mika ; Hyvarinen, Aapo
Author_Institution :
Neural Networks Res. Centre, Helsinki Univ. of Technol., Finland
fDate :
6/24/1905 12:00:00 AM
Abstract :
Estimating overcomplete ICA bases is a difficult problem that emerges when using ICA on many kinds of natural data. Here we introduce two algorithms that estimate an approximate overcomplete basis quite fast in a high-dimensional space. The first algorithm is based on an assumption that the basis vectors are randomly distributed in the space, and the second on the gaussianization procedure
Keywords :
feature extraction; parameter estimation; features; fundamental generative model; gaussianization procedure; image basis vectors; image data; image feature extraction; independent component analysis; overcomplete ICA bases; Feature extraction; Gaussian distribution; Independent component analysis; Matrix decomposition; Neural networks; Space technology; Vectors; Yield estimation;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1005515
