In ordinary independent component analysis, the components are assumed to be completely independent, and they do not necessarily have any meaningful order relationships. In practice, however, the estimated iindependentj components are often not at all independent. We propose that this residual dependence structure could be used to dene a topographic order for the components. In particular, a distance between two components could be dened using their higher-order correlations, and this distance could be used to create a topographic representation. Thus we obtain a linear decomposition into approximately independent components, where the dependence of two components is approximated by the proximity of the components in the topographic representation. 1
|
2147
|
Self-Organizing Maps
– Kohonen
- 2001
|
|
1580
|
A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
– Mallat
- 1989
|
|
940
|
Robust Statistics
– Huber
- 1981
|
|
835
|
Independent component analysis: A new concept
– Comon
- 1994
|
|
710
|
An information-maximization approach to blind separation and blind deconvolution
– Bell, Sejnowski
- 1995
|
|
697
|
Adaptive Filter Theory
– Haykin
- 1996
|
|
442
|
Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381:607–609
– Olshausen, Field
- 1996
|
|
357
|
Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture
– Jutten, Herault
- 1991
|
|
287
|
Equivariant adaptive source separation
– Cardoso, Laheld
- 1996
|
|
283
|
The independent components of natural scenes are edge filters. Vision Research
– Bell, Sejnowski
- 1997
|
|
277
|
What is the goal of sensory coding
– Field
- 1994
|
|
276
|
Blind beamforming for non-Gaussian signals
– Cardoso, Souloumiac
- 1993
|
|
271
|
Probability and Statistics
– DeGroot, Schervish
- 2001
|
|
261
|
Sparse coding with an overcomplete basis set: A strategy employed by V1
– Olshausen, Field
- 1997
|
|
229
|
Principal curves
– Hastie, Stuetzle
- 1989
|
|
227
|
A family of fixed-point algorithms for independent component analysis
– Hyvärinen
- 1997
|
|
226
|
Blind signal separation: Statistical principles
– Cardoso
- 1998
|
|
221
|
Fast and robust fixed-point algorithms for independent component analysis
– Hyvarinen
- 1999
|
|
213
|
GTM: The generative topographic mapping
– Bishop, Svensen, et al.
- 1998
|
|
186
|
Possible Principles Underlying the Transformation of Sensory Messages
– Barlow
- 1961
|
|
174
|
Unsupervised learning
– Barlow
- 1989
|
|
171
|
Exploratory projection pursuit
– Friedman
- 1987
|
|
163
|
Projection pursuit
– Huber
- 1985
|
|
161
|
Independent component filters of natural images compared with simple cells in primary visual cortex
– Hateren, Schaaf
- 1998
|
|
150
|
Normalization of cell responses in cat striate cortex. Visual Neurosci
– Heeger
- 1992
|
|
148
|
Neural networks, principal components, and subspaces
– Oja
- 1989
|
|
140
|
Learning overcomplete representations
– Lewicki, Sejnowski
|
|
138
|
A Projection Pursuit Algorithm for Exploratory Data Analysis
– Friedman, Tukey
- 1974
|
|
122
|
Nonlinear neuron in the low noise limit: A factorial code maximizes information transfer. Network: Computation in neural systems
– Nadal, Parga
- 1994
|
|
116
|
Separation of a mixture of independent signals using time delayed correlations,” Phys
– Molgedey, Schuster
- 1994
|
|
115
|
Independent Component Analysis
– COMON
|
|
114
|
Single units and sensation: A neurone doctrine for perceptual psychology? Perception 1:371–394
– Barlow
- 1972
|
|
114
|
New criteria for blind deconvolution of nonminimum phase systems (channels
– Shalvi, Weinstein
- 1990
|
|
112
|
Garrat,Blind separation of a mixture of independent sources through a quasi-maximum likelihood approach,IEEE Trans. Signal Process
– Pham, P
- 1997
|
|
111
|
Independent component analysis using an extended Infomax algorithm for mixed sub-gaussian and super-gaussian sources
– Lee, Girolami, et al.
- 1999
|
|
107
|
A class of neural networks for independent component analysis
– Karhunen, Oja, et al.
- 1997
|
|
99
|
Sejnowski,Independent component analysis of electroencephalographic data
– Makeig, Bell, et al.
- 1996
|
|
91
|
A method of self-recovering equalization for multi-level amplitude modulation
– Sato
- 1975
|
|
89
|
Adaptive blind separation of independent sources: a deflation approach
– Delfosse, Loubaton
- 1995
|
|
89
|
A context-sensitive generalization of ICA
– Pearlmutter, Parra
- 1996
|
|
81
|
A dimension reduction framework for understanding cortical maps. Nature 343:644–647
– Durbin, Mitchison
- 1990
|
|
78
|
Representations and separation of signals using nonlinear pca type learning, Neural Networks 7
– Karhunen, Joutsensalo
- 1994
|
|
76
|
Robust neural networks with online learning for blind identification and blind separation of sources
– Cichocki, Unbehauen
- 1996
|
|
75
|
Emergence of phase and shift invariant features by decomposition of natural images into independent feature subspaces. Neural Comput
– Hyvarinen, Hoyer
- 2000
|
|
73
|
On stochastic approximation of the eigenvectors and eigenvalues of the expectation of a random matrix
– Oja, Karhunen
- 1985
|
|
72
|
Independent component analysis of natural image sequences yields spatiotemporal filters similar to simple cells in primary visual cortex
– Hateren, Ruderman
- 1998
|
|
70
|
A blind source separation technique based on second order statistics
– Belouchrani, Meraim, et al.
- 1997
|
|
69
|
The development of topography in the visual cortex: a review of models, Network 7
– Swindale
- 1996
|
|
68
|
Sources separation using higher order moments
– Cardoso
- 1989
|
|
66
|
Multi-channel signal separation by decorrelation
– WEINSTEIN, FEDER, et al.
- 1993
|
|
