×

Learned parametric mixture based ICA algorithm. (English) Zbl 0911.68159

Summary: The learned parametric mixture method is presented for a canonical cost function based ICA model on linear mixture, with several new findings. First, its adaptive algorithm is further refined into a simple concise form. Second, the separation ability of this method is shown to be qualitatively superior to its original model with prefixed nonlinearity. Third, a heuristic way is suggested for selecting the number of densities in a learned parametric mixture. Finally, experiments have been conducted to show the success of this method on the sources that can either be sub-Gaussian or super-Gaussian, as well as a combination of both the types.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
68W10 Parallel algorithms in computer science
PDFBibTeX XMLCite
Full Text: DOI Link