Contravariant Adaptation on the Manifold of Causal, FIR, Invertible Multivariable Matrix Systems

This letter extends previous work on contravariant adaptation by providing a formula for the contravariant (natural) gradient on the manifold of multivariable, causal, invertible, finite impulse response (FIR) systems. The right action on the manifold of multivariable causal Toeplitz systems is defined. Using this right action, a bilinear form defined on the tangent space at the identity is extended throughout the entire manifold by invariance. This results in a formula analogous to the natural gradient for matrices, but which preserves the causal, FIR nature of the system. The contravariant conversion factor is block Toeplitz structured, so that implementations may employ fast Fourier transform based convolutions to produce lower complexity than would a comparably-sized generic matrix natural gradient.

[1]  T. Moon,et al.  A Natural Gradient Algorithm for Multichannel Blind Deconvolution: Frequency Domain Criteria and Time Domain Updates , 2006, 2006 IEEE 12th Digital Signal Processing Workshop & 4th IEEE Signal Processing Education Workshop.

[2]  M. T. Vaughn.  Geometry in Physics , 2008 .

[3]  Shun-ichi Amari,et al.  Blind source separation-semiparametric statistical approach , 1997, IEEE Trans. Signal Process..

[4]  Kari Torkkola,et al.  Blind deconvolution, information maximization and recursive filters , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  Andrzej Cichocki,et al.  A New Learning Algorithm for Blind Signal Separation , 1995, NIPS.

[6]  W. Boothby.  An introduction to differentiable manifolds and Riemannian geometry , 1975 .

[7]  S. Amari,et al.  Gradient Learning in Structured Parameter Spaces: Adaptive Blind Separation of Signal Sources , 1996 .

[8]  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.

[9]  Hiroshi Sawada,et al.  Natural gradient multichannel blind deconvolution and speech separation using causal FIR filters , 2004, IEEE Transactions on Speech and Audio Processing.

[10]  Todd K. Moon,et al.  Contravariant Adaptation on the Manifold of Invertible Matrix Transfer Functions , 2017, IEEE Signal Processing Letters.

[11]  Kari Torkkola,et al.  Blind separation of convolved sources based on information maximization , 1996, Neural Networks for Signal Processing VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop.

[12]  Todd K. Moon,et al.  Contravariant adaptation on structured matrix spaces , 2002, Signal Process..

[13]  S.C. Douglas,et al.  Multichannel blind deconvolution of arbitrary signals: adaptive algorithms and stability analyses , 2000, Conference Record of the Thirty-Fourth Asilomar Conference on Signals, Systems and Computers (Cat. No.00CH37154).

[14]  Michael K. Ng,et al.  On inversion of Toeplitz matrices , 2002 .

[15]  Te-Won Lee,et al.  Blind Separation of Delayed and Convolved Sources , 1996, NIPS.

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

[17]  Shun-ichi Amari,et al.  Why natural gradient? , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[18]  Shun-ichi Amari,et al.  Novel On-Line Adaptive Learning Algorithms for Blind Deconvolution Using the Natural Gradient Approach , 1997 .