Kurtosis extrema and identification of independent components: a neural network approach

Mark A. Girolami, Colin Fyfe

Abstract

We propose a nonlinear self-organising network which solely employs computationally simple Hebbian and anti-Hebbian learning in approximating a linear independent component analysis (ICA). Current neural architectures and algorithms which perform parallel ICA are either restricted to positively kurtotic data distributions or data which exhibits one sign of kurtosis . We show that the proposed network is capable of separating mixtures of speech, noise and signals with both platykurtic (positive kurtosis) and leptokurtic (negative kurtosis) distributions in a blind manner. A simulation is reported which successfully separates a mixture of twenty sources of music, speech, noise and fundamental frequencies.
BibTeX
@inproceedings{girolami1997kurtosis,
  author = {Mark A. Girolami and Colin Fyfe},
  title = {{Kurtosis extrema and identification of independent components: a neural network approach}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {1997},
  doi = {10.1109/icassp.1997.595506},
}

Loading related content...

Figure