Representation and separation of signals using nonlinear PCA type learning

Neural Networks

Volumn 7, Issue 1, 1994, Pages 113-127

  Karhunen, Juha ^a   Joutsensalo, Jyrki ^a  

^a AALTO UNIVERSITY   (Finland)

Author keywords

Blind separation; Neural network; Nonlinearity; Oja's rule; Principal components; Signal representation; Sinusoids; Unsupervised learning

Indexed keywords

ALGORITHMS; ESTIMATION; NEURAL NETWORKS; OPTIMIZATION; SIGNAL PROCESSING;

BLIND SEPARATION; SIGNAL REPRESENTATION; SINUSOIDS; UNSUPERVISED LEARNING;

LEARNING SYSTEMS;

EID: 0028272776     PISSN: 08936080     EISSN: None     Source Type: Journal    
DOI: 10.1016/0893-6080(94)90060-4     Document Type: Article

References (38)

1. Back-propagation and unsupervised learning in linear networks
2. Blind separation of sources: A nonlinear neural algorithm
3. Anti-Hebbian learning in a non-linear neural network
4. Blind separation of sources, part II: Problems statement
5. Learning rate schedules for faster stochastic gradient search
6. Forming sparse representations by local anti-Hebbian learning
7. Supporting supervised learning by self-organization
8. Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture
9. On the recursive estimation of the eigenvectors of correlation type matrices
10. Recursive estimation of eigenvectors of correlation type matrices for signal processing applications
11. Tracking of sinusoidal frequencies by neural network learning algorithms
12. Frequency estimation by a Hebbian subspace learning algorithm
13. Nonlinear Hebbian algorithms for sinusoidal frequency estimation
14. Learning of sinusoidal frequencies by nonlinear constrained Hebbian algorithms
15. Hardware implementations of PCA neural networks
16. Neural networks for extracting pure/constrained/oriented principal components
17. Self-organization in a perceptual network
18. A simplified neuron model as a principal component analyzer
19. Neural networks principal components and subspaces
20. Principal components, minor components, and linear neural networks
21. Learning in nonlinear constrained Hebbian networks
22. Principal component analysis by homogeneous neural networks, part II: Analysis and extensions of the learning algorithms
23. An outer product neural network for extracting principal components from a time series
24. Optimal hidden units for two-layer nonlinear feedforward neural networks
25. Unsupervised Hebbian learning and the shape of the neuron activation function
26. A fast neural algorithm for principal component analysis and singular value decomposition
27. Correlations in high dimensional or asymmetric data sets: Hebbian neuronal processing
28. Learning higher order correlations
29. Competitive Hebbian learning: Algorithm and demonstrations
30. Feature discovery through error-correcting learning
31. Least MSE reconstruction for self-organization
32. Least MSE reconstruction for self-organization