Adaptive Blind Signal and Image Processing - Learning Algorithms and Applications

By Andrzej S. Cichocki, Shun-ichi Amari. 2002

1,653 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/164a8b614cdc892c9532d6217ebc371473d4b903

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Diagnosis of Alzheimer's disease from EEG signals: where are we standing? (2010, background): “, [40]), starts from the assumption that EEG signals can be described, to a good approximation, by a finite set of sources, located within the brain; each of those sources generate certain components of the EEG.”
Identifiability, separability, and uniqueness of linear ICA models (2004, background): “Recent textbooks provide excellent tutorial material and extensive review on ICA methods [2], [3].”
Kernel Methods for Measuring Independence (2005, methodology): “…of instantaneous independent component analysis(ICA), drawing on the numerous existing surveys of ICA and related methods, including those by Hyvärinen et al. (2001); Lee et al. (2000); Cichocki and Amari (2002); Haykin (1998); as well as the review by Comon (1994) of older literature on the topic.”
A Fast Approach for Overcomplete Sparse Decomposition Based on Smoothed $\ell ^{0}$ Norm (2008, background): “The goal of Blind Source Separation (BSS) [18], [19] is then to finds(t) only by observingx(t).”
Estimation of a Structural Vector Autoregression Model Using Non-Gaussianity (2010, background): “(3) is, in fact, closely related to the multichannel blind deconvolution problem with causal finite impulse response (FIR) filters (Cichocki and Amari, 2002; Hyv̈arinen et al., 2001).”
Pixel-based and region-based image fusion schemes using ICA bases (2007, methodology): “…ICA rule: a+i ← ai + ηE{z(aTi z)ri}, 1 ≤ i ≤ K (19) A ← A(AT A)−0.5 (20) where η defines the learning rate of the gradient optimisation scheme and ri = K∑ k=1 h(i, k)φ K∑ j=1 h(j, k)(aTi z) 2 (21) As previously, we randomly initialise the update rule in (19) for each projecting vector ai.”
Correlative GC-TOF-MS-based metabolite profiling and LC-MS-based protein profiling reveal time-related systemic regulation of metabolite–protein networks and improve pattern recognition for multiple biomarker selection (2005, methodology): “There is a large variety of methods for performing ICA (Hyvärinen et al., 2001; Cichocki and Amari, 2002).”
Analysis of Sparse Representation and Blind Source Separation (2004, background): “Some learning problems can be expressed in terms of matrix factorization (e.g., Zibulevsky, Pearlmutter, Boll, & Kisilev, 2000; Lee & Seung, 1999; Cichocki & Amari, 2002).”
Tensor Networks for Dimensionality Reduction and Large-scale Optimization: Part 1 Low-Rank Tensor Decompositions (2016, unknown): “The field of 2-way CA is maturing and has generated efficient algorithms for 2-way component analysis, especially for sparse/functional PCA/SVD, ICA, NMF and SCA (Bach and Jordan, 2003; Cichocki and Amari, 2003; Comon and Jutten, 2010; Zhou et al., 2012; Hyvärinen, 2013).”
A Hebbian/Anti-Hebbian Neural Network for Linear Subspace Learning: A Derivation from Multidimensional Scaling of Streaming Data (2015, methodology): “Oja’s rule can be derived by an approximate gradient descent of the mean squared representation error (Cichocki and Amari, 2002; Yang, 1995), a so-called synthesis view of principal component analysis (PCA) (Pearson, 1901; Preisendorfer and Mobley, 1988): (2) Computing principal components beyond…”
Information theoretical estimators toolbox (2014, background): “…multi-information when two variables are considered. c©2014 Zoltán Szabó. the ISA separation principle (Cardoso, 1998): the ISA task can be solved by ICA (ISA with one-dimensional sources, Hyvärinen et al. 2001; Cichocki and Amari 2002; Choi et al. 2005) followed by clustering of the ICA elements.”
Constrained Nonnegative Matrix Factorization for Hyperspectral Unmixing (2009, background): “The other group of approaches considers the unsupervised linear hyperspectral unmixing as a blind source separation (BSS) problem [11].”