PuSH - Publikationsserver des Helmholtz Zentrums München

Information-theoretic model selection for independent components.

In: Proceedings (Latent variable analysis and signal separation : 9th international conference, 27-30 September 2010, St. Malo, France). Berlin [u.a.]: Springer, 2010. 254-262 (Lecture Notes Comp. Sci. ; 6365)
Independent Component Analysis (ICA) is an essential building block for data analysis in many applications. Selecting the truly meaningful components from the result of an ICA algorithm, or comparing the results of different algorithms, however, are non-trivial problems. We introduce a very general technique for evaluating ICA results rooted in information-theoretic model selection. The basic idea is to exploit the natural link between non-Gaussianity and data compression: The better the data transformation represented by one or several ICs improves the effectiveness of data compression, the higher is the relevance of the ICs. In an extensive experimental evaluation we demonstrate that our novel information-theoretic measure robustly selects the most interesting components from data without requiring any assumptions or thresholds.
Altmetric
Weitere Metriken?
Zusatzinfos bearbeiten [➜Einloggen]
Publikationstyp Artikel: Sammelbandbeitrag/Konferenzbeitrag
ISSN (print) / ISBN 0302-9743
e-ISSN 1611-3349
ISBN 364215994X
Konferenztitel Latent variable analysis and signal separation : 9th international conference
Konferzenzdatum 27-30 September 2010
Konferenzort St. Malo, France
Konferenzband Proceedings
Zeitschrift Lecture Notes in Computer Science
Quellenangaben Band: 6365, Heft: , Seiten: 254-262 Artikelnummer: , Supplement: ,
Verlag Springer
Verlagsort Berlin [u.a.]
Begutachtungsstatus