Abstract
In this paper, I introduce a concept of differential decorrelation which finds a linear mapping that minimizes the concurrent change of variables. Motivated by the differential anti-Hebbian rule [1], I develop a natural gradient algorithm for differential decorrelation and present its local stability analysis. The algorithm is successfully applied to the task of nonstationary source separation.
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.
Similar content being viewed by others
References
Choi, S.: Differential Hebbian-type learning algorithms for decorrelation and independent component analysis. Electronics Letters 34 (1998) 900–901
Hebb, D.O.: The Organization of Behavior. Wiley, New York (1949)
Földiák, P.: Adaptive network for optimal linear feature extraction. In: Proc. Int. Joint Conf. Neural Networks. (1989) 401–405
Kosko, B.: Differential Hebbian learning. In: Proc. American Institute of Physics: Neural Networks for Computing. (1986) 277–282
Amari, S.: Natura1 gradient works efficiently in learning. Neural Computation 10 (1998) 251–276
Choi, S., Cichocki, A., Amari, S.: Equivariant nonstationary source separation. Neural Networks 15 (2002) 121–130
Amari, S., Chen, T.P., Cichocki, A.: Stability analysis of learning algorithms for blind source separation. Neural Networks 10 (1997) 1345–1351
Choi, S., Cichocki, A., Belouchrani, A.: Second order nonstationary source separation. Journal of VLSI Signal Processing (2002) to appear.
Matsuoka, K., Ohya, M., Kawamoto, M.: A neural net for blind Separation of nonstationary signals. Neural Networks 8 (1995) 411–419
Choi, S., Cichocki, A., Deville, Y.: Differential decorrelation for nonstationary source separation. In: Proc. ICA, San Diego, CA (2001)
Deville, Y., Benali, M.: Differential source separation: Concept and application to a criterion based on differential normalized kurtosis. In: Proc. EUSIPCO. (2000)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Choi, S. (2002). Adaptive Differential Decorrelation: A Natural Gradient Algorithm. In: Dorronsoro, J.R. (eds) Artificial Neural Networks — ICANN 2002. ICANN 2002. Lecture Notes in Computer Science, vol 2415. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46084-5_189
Download citation
DOI: https://doi.org/10.1007/3-540-46084-5_189
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44074-1
Online ISBN: 978-3-540-46084-8
eBook Packages: Springer Book Archive