Abstract
Recently, Plumbley and Oja presented a non-negative PCA algorithm, which performs blind separation of non-negative well-grounded sources using a rectification nonlinearity. The algorithm is based on an ordinary gradient learning, and the observations require prewhitening by eigenvalue decomposition. In this paper, we apply our previously developed natural-gradient-based RLS algorithm to optimize the non-negative PCA criterion. In addition, we propose here an RLS-type preprocessing algorithm, which can whiten the data in terms of covariance matrix. These two algorithms are adaptive, and can work in a cascade mode. The validity of the new implementation is confirmed through computer simulations.
This work was supported in part by the Chinese Postdoctoral Science Foundation and in part by the Foundation of Intel China Research Center.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Hyvarinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. Wiley, New York (2001)
Plumbley, M.: Algorithms for Nonnegative Independent Component Analysis. IEEE Trans. Neural Networks 14, 534–543 (2003)
Plumbley, M., Oja, E.: A Nonnegative PCA Algorithms for Independent Component Analysis. IEEE Trans. Neural Networks 15, 66–76 (2004)
Oja, E.: The Nonlinear PCA Learning in Independent Component Analysis. Neurocomputing 17, 25–46 (1997)
Haykin, S.: Adaptive Filter Theory, 3rd edn. Prentice-Hall, Englewood Cliffs (1996)
Karhunen, J., Pajunen, P., Oja, E.: The Nonlinear PCA Criterion in Blind Source Separation: Relations with Other Approaches. Neurocomputing 22, 5–20 (1998)
Amari, S.: Natural Gradient Works Efficiently in Learning. Neural Computation 10, 251–276 (1998)
Amari, S.: Natural Gradient Learning for Over- and Under-Complete Bases in ICA. Neural Computation 11, 1875–1883 (1999)
Zhu, X.L., Zhang, X.D.: Adaptive RLS Algorithm for Blind Source Separation Using a Natural Gradient. IEEE Signal Processing Letters 9, 432–435 (2002)
Lee, D.D., Seung, H.S.: Learning the Parts of Objects by Nonnegative Matrix Factorization. Nature 401, 788–791 (1999)
Cardoso, J.F., Laheld, H.: Equivariant Adaptive Source Separation. IEEE Trans. Signal Processing 44, 3017–3029 (1996)
Zhu, X.L., Zhang, X.D., Ye, J.M.: Natural Gradient-Based Recursive Least-Squares Algorithm for Adaptive Blind Source Separation. Science in China Ser. F 47, 55–65 (2004)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Zhu, XL., Zhang, XD., Jia, Y. (2004). Adaptive RLS Implementation of Non-negative PCA Algorithm for Blind Source Separation. In: Yin, FL., Wang, J., Guo, C. (eds) Advances in Neural Networks – ISNN 2004. ISNN 2004. Lecture Notes in Computer Science, vol 3173. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-28647-9_125
Download citation
DOI: https://doi.org/10.1007/978-3-540-28647-9_125
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-22841-7
Online ISBN: 978-3-540-28647-9
eBook Packages: Springer Book Archive