Date: January 7 2021
Summary: What independent component analysis is
Keywords: ##zettel #mathematics #analysis #independent #component #analysis #speech #fmri #noise #separation #archive
Not Available
ICA seeks to minimize mutual information between projections.Independent component analysis finds a coordinate frame onto which the projection of the data has minimal temporal overlap. [1]
ICA is considered a linear decomposition alternative to principal component analysis (PCA). In PCA, the data is represented using perpendicular axes. ICA is not limited by this constraint [2]
ICA has been applied to domains such as:
fMRI
performing speech
noise separation [3]
ICA decompositions are best for sources linearly mixed in a recorded signal.
Zelko, Jacob. Independent Component Analysis. https://jacobzelko.com/01072021054854-independent-component-analysis. January 7 2021.
[1] A. Delorme, S. Makeig, M. Fabre-Thorpe, and T. Sejnowski, “From single-trial EEG to brain area dynamics,” Neurocomputing, vol. 44–46, pp. 1057–1064, Jun. 2002, doi: 10.1016/S0925-2312(02)00415-0.
[2] S. Makeig et al., “Functionally independent components of the late positive event-related potential during visual spatial attention,” J. Neurosci., vol. 19, no. 7, pp. 2665–2680, 1999.
[3] H.-M. Park, H.-Y. Jung, T.-W. Lee, and S.-Y. Lee, “Subband-based blind signal separation for noisy speech recognition,” Electron. Lett., vol. 35, no. 23, pp. 2011–2012, 1999.