TY - JOUR
T1 - Evaluating model misspecification in independent component analysis
AU - Lee, Seonjoo
AU - Caffo, Brian S.
AU - Lakshmanan, Balaji
AU - Pham, Dzung L.
N1 - Funding Information:
The project described was partially supported by grants R01-EB012547 and P41-EB015909 from the National Institute of Biomedical Imaging and Bioengineering, and R01-NS070906 from the National Institute of Neurological Disorders and Stroke.
Publisher Copyright:
© 2013, © 2013 Taylor & Francis.
PY - 2015/4/13
Y1 - 2015/4/13
N2 - Independent component analysis (ICA) is a popular blind source separation technique used in many scientific disciplines. Current ICA approaches have focused on developing efficient algorithms under specific ICA models, such as instantaneous or convolutive mixing conditions, intrinsically assuming temporal independence or autocorrelation of the sources. In practice, the true model is not known and different ICA algorithms can produce very different results. Although it is critical to choose an ICA model, there has not been enough research done on evaluating mixing models and assumptions, and how the associated algorithms may perform under different scenarios. In this paper, we investigate the performance of multiple ICA algorithms under various mixing conditions. We also propose a convolutive ICA algorithm for echoic mixing cases. Our simulation studies show that the performance of ICA algorithms is highly dependent on mixing conditions and temporal independence of the sources. Most instantaneous ICA algorithms fail to separate autocorrelated sources, while convolutive ICA algorithms depend highly on the model specification and approximation accuracy of unmixing filters.
AB - Independent component analysis (ICA) is a popular blind source separation technique used in many scientific disciplines. Current ICA approaches have focused on developing efficient algorithms under specific ICA models, such as instantaneous or convolutive mixing conditions, intrinsically assuming temporal independence or autocorrelation of the sources. In practice, the true model is not known and different ICA algorithms can produce very different results. Although it is critical to choose an ICA model, there has not been enough research done on evaluating mixing models and assumptions, and how the associated algorithms may perform under different scenarios. In this paper, we investigate the performance of multiple ICA algorithms under various mixing conditions. We also propose a convolutive ICA algorithm for echoic mixing cases. Our simulation studies show that the performance of ICA algorithms is highly dependent on mixing conditions and temporal independence of the sources. Most instantaneous ICA algorithms fail to separate autocorrelated sources, while convolutive ICA algorithms depend highly on the model specification and approximation accuracy of unmixing filters.
KW - convolutive mixing
KW - independent component analysis
KW - model misspecification
UR - http://www.scopus.com/inward/record.url?scp=84921416027&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84921416027&partnerID=8YFLogxK
U2 - 10.1080/00949655.2013.867961
DO - 10.1080/00949655.2013.867961
M3 - Article
C2 - 25642002
AN - SCOPUS:84921416027
SN - 0094-9655
VL - 85
SP - 1151
EP - 1164
JO - Journal of Statistical Computation and Simulation
JF - Journal of Statistical Computation and Simulation
IS - 6
ER -