Approximate Maximum Likelihood Source Separation Using the Natural Gradient(Regular Section)
スポンサーリンク
概要
- 論文の詳細を見る
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.
- 社団法人電子情報通信学会の論文
- 2003-01-01
著者
-
CICHOCKI Andrzej
Brain Science Institute, RIKEN
-
AMARI Shun-ichi
Brain-Style Information Systems Group, Brain Science Institute, RIKEN
-
CHOI Seungjin
Department of Computer Science and Engineering, Pohang University of Science and Technology
-
Cichocki Andrzej
Brain Science Institute Riken
-
Choi Seungjin
Department Of Computer Science And Engineering
-
Amari Shun-ichi
Brain-style Information Systems Research Group
-
Cichocki Andrzej
Brain-style Information Systems Research Group
-
ZHANG Liqing
Brain-style Information Systems Research Group
-
Zhang L
Brain-style Information Systems Research Group
-
Amari Shun-ichi
Brain Science Institute Riken Brain-style Information Systems Research Group
関連論文
- Saccade-related independent component in visually and auditorily guided saccade task
- Single-Trial Magnetoencephalographic Data Decomposition and Localization Based on Independent Component Analysis Approach
- Neural Network Models for Blind Separation of Time Delayed and Convolved Signals
- Equivariant nonstationary source separation
- Natural Gradient Learning for Spatio-Temporal Decorrelation:Recurrent Network
- Blind Identification and Separation of Noisy Source Signals : Neural Network Approaches
- A Cascade Neural Network for Blind Signal Extraction without Spurious Equilibria (Special Section on Nonlinear Theory and Its Applications)
- Approximate Maximum Likelihood Source Separation Using the Natural Gradient(Regular Section)
- 3D Object Recognition System Based on 2D Chain Code Matching (Special Section on JTC-CSCC '92)
- Tensor decompositions for feature extraction and classification of high dimensional datasets
- Differential learning algorithms for decorrelation and independent component analysis
- Prediction of post-translational phosphorylation of proteins using multiple kernel learning(Biometrics1)