This paper addresses a maximum likelihood method for source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We consider an approximate likelihood which is based on the Laplace approximation and develop a natural gradient adaptation algorithm to find a local maximum of the corresponding approximate likelihood. We present a detailed mathematical derivation of the algorithm using the Lie group invariance. Useful behavior of the algorithm is verified by numerical experiments.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Seungjin CHOI, Andrzej CICHOCKI, Liqing ZHANG, Shun-ichi AMARI, "Approximate Maximum Likelihood Source Separation Using the Natural Gradient" in IEICE TRANSACTIONS on Fundamentals,
vol. E86-A, no. 1, pp. 198-205, January 2003, doi: .
Abstract: This paper addresses a maximum likelihood method for source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We consider an approximate likelihood which is based on the Laplace approximation and develop a natural gradient adaptation algorithm to find a local maximum of the corresponding approximate likelihood. We present a detailed mathematical derivation of the algorithm using the Lie group invariance. Useful behavior of the algorithm is verified by numerical experiments.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e86-a_1_198/_p
Copy
@ARTICLE{e86-a_1_198,
author={Seungjin CHOI, Andrzej CICHOCKI, Liqing ZHANG, Shun-ichi AMARI, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Approximate Maximum Likelihood Source Separation Using the Natural Gradient},
year={2003},
volume={E86-A},
number={1},
pages={198-205},
abstract={This paper addresses a maximum likelihood method for source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We consider an approximate likelihood which is based on the Laplace approximation and develop a natural gradient adaptation algorithm to find a local maximum of the corresponding approximate likelihood. We present a detailed mathematical derivation of the algorithm using the Lie group invariance. Useful behavior of the algorithm is verified by numerical experiments.},
keywords={},
doi={},
ISSN={},
month={January},}
Copy
TY - JOUR
TI - Approximate Maximum Likelihood Source Separation Using the Natural Gradient
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 198
EP - 205
AU - Seungjin CHOI
AU - Andrzej CICHOCKI
AU - Liqing ZHANG
AU - Shun-ichi AMARI
PY - 2003
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E86-A
IS - 1
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - January 2003
AB - This paper addresses a maximum likelihood method for source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We consider an approximate likelihood which is based on the Laplace approximation and develop a natural gradient adaptation algorithm to find a local maximum of the corresponding approximate likelihood. We present a detailed mathematical derivation of the algorithm using the Lie group invariance. Useful behavior of the algorithm is verified by numerical experiments.
ER -