Neural network implementations of independent component analysis
Tóm tắt
The performance of six neuromorphic adaptive structurally different algorithms was analyzed in blind separation of independent artificially generated signals using the stationary linear independent component analysis (ICA) model. The estimated independent components were assessed and compared aiming to rank the neural ICA implementations. All algorithms were run with different contrast functions, which were optimally selected on the basis of maximizing the sum of individual negentropies of the network outputs. Both subGaussian and superGaussian one-dimensional time series were employed throughout the numerical simulations.
Từ khóa
#Neural networks #Independent component analysis #Signal processing algorithms #Principal component analysis #Source separation #Array signal processing #Vectors #Artificial neural networks #Psychology #NeuromorphicsTài liệu tham khảo
10.1162/neco.1997.9.7.1483
10.1049/ip-f-2.1993.0054
pham, 1992, Separation of a mixture of independent sources through a maximum likelihood approach, Proc EUSIPCO, 771
10.1162/neco.1997.9.7.1457
10.1080/01621459.1987.10478427
lee, 2000, A unifying information-theoretic framework for Independent Component Analysis, Computers and Mathematics with Applications, 39, 1
hyvärinen, 1998, New approximations of differential entropy for ICA and projection pursuit, Advances in Neural Information Processing Systems (NIPS'97), 10, 273
abed meraim, 1994, Asymptotic performance of second order blind source separation, Proc IEEE ICASSP IV, 277
10.1016/0165-1684(91)90079-X
10.1162/neco.1995.7.6.1129
10.1109/HOST.1993.264564
amari, 1996, A new learning algorithm for blind source separation, Advances in Neural Information Processing (Proc NIPS'95)
10.1109/5.720250
10.1016/0165-1684(94)90029-9
10.1109/ICASSP.1994.389822