A comparison of algorithms for separation of synchronous subspaces

• Autorzy: Almeida M. , Bioucas-Dias J. , Vigário R. , Oja E.
• Języki publikacji: EN

Abstrakty

EN

Independent Subspace Analysis (ISA) consists in separating sets (subspaces) of dependent sources, with different sets being independent of each other. While a few algorithms have been proposed to solve this problem, they are all completely general in the sense that they do not make any assumptions on the intra-subspace dependency. In this paper, we address the ISA problem in the specific context of Separation of Synchronous Sources (SSS), i.e., we aim to solve the ISA problem when the intra-subspace dependency is known to be perfect phase synchrony between all sources in that subspace. We compare multiple algorithmic solutions for this problem, by analyzing their performance on an MEG-like dataset.

Słowa kluczowe

EN

phase locking; synchrony; source separation; subspaces; independent component analysis; ICA; Independent Subspace Analysis (ISA); magnetoencephalogram (MEG)

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2012
• Tom: Vol. 60, nr 3
• Strony: 455--460
• Opis fizyczny: Bibliogr. 26 poz., rys., tab.

Twórcy

autor

Almeida M.

autor

Bioucas-Dias J.

autor

Vigário R.

autor

Oja E.

• Instituto de Telecomunicac¸ ~oes, Instituto Superior T´ecnico, Technical University of Lisbon, Av. Rovisco Pais, 1049-001 Lisbon, Portugal / Department of Information and Computer Science, Aalto University, F1-00076 Aalto, Finland

Bibliografia

• [1] A. Pikovsky, M. Rosenblum, and J. Kurths, Synchronization: aUniversal Concept in Nonlinear Sciences, Cambridge NonlinearScience Series, Cambridge University Press, Cambridge, 2001.
• [2] P.J. Uhlhaas and W. Singer, “Neural synchrony in brain disorders: relevance for cognitive dysfunctions and pathophysiology”, Neuron 52, 155-168 (2006).
• [3] P.L. Nunez, R. Srinivasan, A.F. Westdorp, R.S. Wijesinghe, D.M. Tucker, R.B. Silberstein, and P.J. Cadusch, “EEG coherency I: statistics, reference electrode, volume conduction, laplacians, cortical imaging, and interpretation at multiple scales”, Electroencephalography and Clinical Neurophysiology 103, 499-515 (1997).
• [4] R. Vigário, J. Särelä, V. Jousmäki, M. Hämäläinen, and E. Oja, “Independent component approach to the analysis of EEG and MEG recordings”, IEEE Trans. on Biom. Eng. 47 (5), 589-593 (2000).
• [5] A. Hyvärinen, J. Karhunen, and E. Oja, Independent ComponentAnalysis. John Wiley & Sons, London, 2001.
• [6] A. Cichocki and S. Amari, Adaptive Blind Signal and ImageProcessing - Learning Algorithms and Applications, John Wiley & Sons, London, 2002.
• [7] P. Hoyer, “Non-negative matrix factorization with sparseness constraints”, J. Machine Learning Research 5, 1457-1469 (2004).
• [8] D. Lee and H. Seung, “Algorithms for non-negative matrix factorization”, Advances in Neural Information Processing Systems 13, 556-562 (2001).
• [9] M. Almeida, J.-H. Schleimer, J. Bioucas-Dias, and R. Vigärio, “Source separation and clustering of phase-locked subspaces”, IEEE Trans. on Neural Networks 22 (9), 1419-1434 (2011).
• [10] S. Boyd and L. Vandenberghe, Convex Optimization, Cambridge University Press, Cambridge, 2004.
• [11] E. Alhoniemi, A. Honkela, K. Lagus, and J. Seppä, “Compact modeling of data using independent variable group analysis”, IEEE Trans. on Neural Networks 18, 1762-1776 (2007).
• [12] B. Póczos and A. Lörincz, “Independent subspace analysis using geodesic spanning trees”, Proc. Int. Conf. on MachineLearning (ICML) 1, CD-ROM (2005).
• [13] J. Beirlant, E. Dudewicz, L. Gyorfi, and E. van der Meulen, “Nonparametric entropy estimation: an overview”, Int. J. Mathematicaland Statistical Sciences 6, 17-39 (1997).
• [14] H.W. Gutch, J. Krumsiek, and F.J. Theis, “An ISA algorithm with unknown group sizes identifies meaningful clusters in metabolomics data”, Proc. Eur. Signal Processing Conf. (EUSIPCO) 1, CD-ROM (2011).
• [15] A. Hyvärinen and U. Köster, “FastISA: A fast fixed-point algorithm for independent subspace analysis”, Proc. Eur. Symposiumon Artificial Neural Networks (ESANN) 1, CD-ROM (2006).
• [16] A. Sharma and K.K. Paliwal, “Subspace independent component analysis using vector kurtosis”, Pattern Recognition 39, 2227-2232 (2006).
• [17] J.A. Palmer and S. Makeig, “Blind separation of dependent sources and subspaces by minimum mutual information”, in Technical Report, University of California, San Diego, 2010.
• [18] Z. Szabó, B. Póczos, and A. Lorincz, “Undercomplete blind subspace deconvolution”, J. Machine Learning Research 8, 1063-1095 (2007).
• [19] F.J. Theis, “Towards a general independent subspace analysis”, Advances in Neural Information Processing Systems (NIPS) 1, CD-ROM (2007).
• [20] L.B. Almeida, “MISEP - linear and nonlinear ICA based on mutual information”, J. Machine Learning Research 4, 1297- 1318 (2004).
• [21] A. Ziehe and K.-R. Müller, “TDSEP - an efficient algorithm for blind separation using time structure”, Int. Conf. on ArtificialNeural Networks 1, 675-680 (1998).
• [22] R. Vigário, V. Jousmäki, M. Hämäläinen, R. Hari, and E. Oja, “Independent component analysis for identification of artifacts in magnetoencephalographic recordings”, Advances in NIPS 10, 229-235 (1997).
• [23] A.V. Oppenheim, R.W. Schafer, and J.R. Buck, Discrete-TimeSignal Processing, Prentice-Hall International Editions, London, 1999.
• [24] T. Eichele, S. Rachakonda, B. Brakedal, R. Eikeland, and V. D. Calhoun, “EEGIFT: Group independent component analysis for event-related EEG data”, Computational Intelligence andNeuroscience 2011, 1-9 (2011).
• [25] M. Almeida, R. Vigário, and J. Bioucas-Dias, “Estimation of the common oscillation for phase locked matrix factorization”, Proc. Int. Conf. on Pattern Recognition Applications andMethods (ICPRAM) 1, CD-ROM (2012).
• [26] P.L. Nunez and R. Srinivasan, Electric Fields of the Brain: theNeurophysics of EEG, Oxford University Press, Oxford, 2006