Abstract
Linear multilayer independent component analysis (LMICA) is an approximate algorithm for ICA. In LMICA, approximate independent components are efficiently estimated by optimizing only highly dependent pairs of signals when all the sources are super-Gaussian. In this paper, the nonlinear functions in LMICA are generalized, and a new method using adaptive PCA is proposed for the selection of pairs of highly dependent signals. In this method, at first, all the signals are sorted along the first principal axis of their higher-order correlation matrix. Then, the sorted signals are divided into two groups so that relatively highly correlated signals are collected in each group. Lastly, each of them is sorted recursively. This process is repeated until each group consists of only one or two signals. Because a well-known adaptive PCA algorithm named PAST is utilized for calculating the first principal axis, this method is quite simple and efficient. Some numerical experiments verify the effectiveness of LMICA with this improvement.
Similar content being viewed by others
References
Amari S (1998) Natural gradient works efficiently. Neural Comput 10: 251–276
Amari S, Cichocki A (1996) A new learning algorithm for blind signal separation. In: Touretzky D, Mozer M, Hasselmo M (eds) Advances in neural information processing systems 8. MIT Press, Cambridge, MA, pp 757–763
Bell AJ, Sejnowski TJ (1997) The independent components of natural scenes are edge filters. Vision Res 37(23): 3327–3338
Burnham KP, Anderson DR (2002) Model selection and multimodel inference: a practical-theoretic approach, 2nd edn. Springer, Berlin, NY
Cardoso J-F (1999) High-order contrasts for independent component analysis. Neural Comput 11(1): 157–192
Cichocki A, Amari S (2002) Adaptive blind signal and image processing: learning algorithms and applications. Wiley, London
Cichocki A, Kasprzak W, Amari S (1995) Multi-layer neural networks with a local adaptive learning rule for blind separation of source signals. In: Proceedings of NOLTA95, vol 1. Las Vegas, USA, pp 61–66
Cichocki A, Zdunek R (2007) Multilayer nonnegative matrix factorization using projected gradient approaches. Int J Neural Syst 17(6): 431–446
Hyvärinen A (1999) Fast and robust fixed-point algorithms for independent component analysis. IEEE Trans Neural Netw 10(3): 626–634
Hyvärinen A, Karhunen J, Oja E (2001) Independent component analysis. Wiley, London
Matsuda Y, Yamaguchi K (2005) An efficient MDS-based topographic mapping algorithm. Neurocomputing 64: 285–299
Matsuda Y, Yamaguchi K (2007) Extraction of approximate independent components from large natural scenes. In: ICONIP 2007, vol 4984. Springer, Kitakyushu, pp 635–642
Matsuda Y, Yamaguchi K (2007) Linear multilayer ICA generating hierarchical edge detectors. Neural Comput 19: 218–230
Matsuda Y, Yamaguchi K (2008) A connection-limited neural network by infomax and infomin. In: Proceedings of IJCNN2008, Hong Kong, IEEE, pp 2531–2537
Oja E (1982) A simplified neuron model as a principal component analyzer. J Math Biol 15: 267–273
Oja E (1992) Principal components, minor components, and linear neural networks. Neural Netw 5: 927–935
Olshausen BA, Field DJ (1996) Emergence of simple-cell receptive field properties by learning a sparce code for natural images. Nature 381: 607–609
Ouyang S, Bao Z, Liao G-S (2000) Robust recursive least squares learning algorithm for principal component analysis. IEEE Trans Neural Netw 11(1): 215–221
Sanger TD (1989) Optimal unsupervised learning in a single-layer linear feedforward neural network. Neural Netw 2: 459–473
Särelä J, Vigário R (2003) Overlearning in marginal distribution-based ica: analysis and solutions. J Mach Learn Res 4: 1447–1469
van Hateren JH, van der Schaaf A (1998) Independent component filters of natural images compared with simple cells in primary visual cortex. Proc Royal Soc Lond B 265: 359–366
Yang B (1995) Projection approximation subspace tracking. IEEE Trans Signal Process 43(1): 95–107
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Matsuda, Y., Yamaguchi, K. Linear Multilayer ICA Using Adaptive PCA. Neural Process Lett 30, 133–144 (2009). https://doi.org/10.1007/s11063-009-9114-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-009-9114-4