Thumbnail
Access Restriction
Open

Author Himberg, Johan
Source CiteSeerX
Content type Text
File Format PDF
Language English
Subject Domain (in DDC) Computer science, information & general works ♦ Data processing & computer science
Subject Keyword Independent Component Analysis ♦ Strong Noise ♦ Mixing Model ♦ Low-noise As ♦ Order Umu-lants ♦ Binary Op-erations ♦ Experimental Study ♦ Independent Binary Omponents ♦ Sparse Data ♦ Extensive Simulation ♦ Order Umulant Performs ♦ Binary Data ♦ Ordinary Umulant-based Ica Algorithm
Description Workshop Independent Component Analysis and Blind Signal Separation
We onsider a mixing model where independent binary omponents are mixed using binary OR op-erations. Using extensive simulations, we investigate whether the model an be estimated using ordinary umulant-based ICA algorithms. We show that the model an indeed be estimated if the data is sparse enough. We also ompare the 3rd and 4th order umu-lants. In the no-noise and low-noise ases, the 3rd order umulant performs better, but in the presen e of strong noise, the 4th-order umulant, somewhat surprisingly, performs better for very sparse data. 1.
Educational Role Student ♦ Teacher
Age Range above 22 year
Educational Use Research
Education Level UG and PG ♦ Career/Technical Study
Learning Resource Type Article
Publisher Date 2001-01-01