论文信息 - An Experimental Comparison of Neural ICA Algorithms

An Experimental Comparison of Neural ICA Algorithms

Several neural algorithms for Independent Component Analysis (ICA) have been introduced lately, but their computational properties have not yet been systematically studied. In this paper, we compare the accuracy, convergence speed, computational load, and other properties of five prominent neural or semi-neural ICA algorithms. The comparison reveals some interesting differences between the algorithms.

[1] J. Friedman. Exploratory Projection Pursuit , 1987 .

[2] M. V. Rossum,et al. In Neural Computation , 2022 .

[3] Christian Jutten,et al. Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[4] Pierre Comon,et al. Independent component analysis, A new concept? , 1994, Signal Process..

[5] Terrence J. Sejnowski,et al. An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[6] Jean-François Cardoso,et al. Equivariant adaptive source separation , 1996, IEEE Trans. Signal Process..

[7] Andrzej Cichocki,et al. Robust neural networks with on-line learning for blind identification and blind separation of sources , 1996 .

[8] Erkki Oja,et al. A class of neural networks for independent component analysis , 1997, IEEE Trans. Neural Networks.

[9] Shun-ichi Amari,et al. Adaptive Online Learning Algorithms for Blind Separation: Maximum Entropy and Minimum Mutual Information , 1997, Neural Computation.

[10] Nikola Kasabov,et al. Brain-like Computing and Intelligent Information Systems , 1998 .

[11] S Makeig,et al. Spatially independent activity patterns in functional MRI data during the stroop color-naming task. , 1998, Proceedings of the National Academy of Sciences of the United States of America.