Natural Gradient Learning for Spatio-temporal Decorrelation : Recurrent Network

Spatio-temporal decorrelation is the task of eliminating correlations between associated signals in spatial domain as well as in time domain. In this paper, we present a simple but eÆcient adaptive algorithm for spatio-temporal decorrelation. For the task of spatio-temporal decorrelation, we consider a dynamic recurrent network and calculate the associated natural gradient for the minimization of an appropriate optimization function. The natural gradient based spatio-temporal decorrelation algorithm is applied to the task of blind deconvolution of linear single input multiple output (SIMO) system and its performance is compared to the spatio-temporal anti-Hebbian learning rule.

[1]  Shun-ichi Amari,et al.  Blind equalization of SIMO channels via spatio-temporal anti-Hebbian learning rule , 1998, Neural Networks for Signal Processing VIII. Proceedings of the 1998 IEEE Signal Processing Society Workshop (Cat. No.98TH8378).

[2]  Seungjin Choi,et al.  Natural Gradient Learning for Spatio-Temporal Decorrelation: Recurrent Network , 2000 .

[3]  Lang Tong,et al.  A new approach to blind identification and equalization of multipath channels , 1991, [1991] Conference Record of the Twenty-Fifth Asilomar Conference on Signals, Systems & Computers.

[4]  Seungjin Choi Adaptive Blind Equalization for Multichannel Transmission via Temporal Decorrelation , 1999 .

[5]  Jie Zhu,et al.  Anti-Hebbian learning in topologically constrained linear networks: a tutorial , 1993, IEEE Trans. Neural Networks.

[6]  Andrzej Cichocki,et al.  An unsupervised hybrid network for blind separation of independent non-Gaussian source signals in multipath environment , 1999, Journal of Communications and Networks.

[7]  S.C. Douglas,et al.  Multichannel blind deconvolution and equalization using the natural gradient , 1997, First IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications.

[8]  Shun-ichi AMARIyy,et al.  NATURAL GRADIENT LEARNING WITH A NONHOLONOMIC CONSTRAINT FOR BLIND DECONVOLUTION OF MULTIPLE CHANNELS , 1999 .

[9]  P. Foldiak,et al.  Adaptive network for optimal linear feature extraction , 1989, International 1989 Joint Conference on Neural Networks.

[10]  S. Choi,et al.  An adaptive system for direct blind multi-channel equalization , 1997, First IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications.

[11]  Seungjin Choi Linear Neural Networks with FIR Synapses for Blind Deconvolution and Equalization , 1999 .

[12]  Erkki Oja,et al.  Neural Networks, Principal Components, and Subspaces , 1989, Int. J. Neural Syst..

[13]  Dirk Van Compernolle,et al.  Signal separation by symmetric adaptive decorrelation: stability, convergence, and uniqueness , 1995, IEEE Trans. Signal Process..

[14]  Shun-ichi Amari,et al.  Natural Gradient Works Efficiently in Learning , 1998, Neural Computation.

[15]  Andrzej Cichocki,et al.  Natural Gradient Learning Algorithms for Decorrelation , 1997, ICONIP.