Abstract
In this paper, we propose a new algorithm for a flexible group multi-way data analysis called the linked Tucker2 decomposition (LT2D). The LT2D can decompose given multiple tensors into common factor matrices, individual factor matrices, and core tensors, simultaneously. When we have a set of tensor data and want to estimate common components and/or individual characteristics of the data, this decomposition model is very useful. In order to develop an efficient algorithm for the LT2D, we imposed orthogonality constraints to factor matrices and applied alternating least squares (ALS) algorithm to the optimization criterion. We conducted some experiments to demonstrate the advantages and convergence properties of the proposed algorithm. Finally, we discuss potential applications of the proposed method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Amari, S., Cichocki, A., Yang, H.H.: A new learning algorithm for blind signal separation. In: Advances in Neural Information Processing Systems, pp. 757–763 (1996)
Carroll, J., Chang, J.J.: Analysis of individual differences in multidimensional scaling via an n-way generalization of ‘Eckart-Young’ decomposition. Psychometrika 35, 283–319 (1970)
Cichocki, A., Zdunek, R., Phan, A.H., Amari, S.: Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source Separation. Wiley Publishing (2009)
Crainiceanu, C.M., Caffo, B.S., Sheng, L., Zipunnikov, V.M., Punjabi, N.M.: Population value decomposition, a framework for the analysis of image populations. Journal of the American Statistical Association 106(495), 775–790 (2011)
Geladi, P., Kowalski, B.R.: Partial least-squares regression: A tutorial. Analytica Chimica Acta 185, 1–17 (1986)
Jolliffe, I.T.: Principal component analysis, vol. 487. Springer, New York (1986)
Jolliffe, I.T., Trendafilov, N.T., Uddin, M.: A modified principal component technique based on the lasso. Journal of Computational and Graphical Statistics 12(3), 531–547 (2003)
Kowalski, B., Gerlach, R., Wold, H.: Chemical Systems under Indirect Observation. In: Systems Under Indirect Observation, pp. 191–209. North-Holland, Amsterdam (1986)
Kroonenberg, P.M., De Leeuw, J.: Principal component analysis of three-mode data by means of alternating least squares algorithms. Psychometrika 45(1), 69–97 (1980)
Lock, E.F., Hoadley, K.A., Marron, J., Nobel, A.B.: Joint and individual variation explained (JIVE) for integrated analysis of multiple data types. The Annals of Applied Statistics 7(1), 523 (2013)
Phan, A., Cichocki, A.: Tensor decompositions for feature extraction and classification of high dimensional datasets. Nonlinear Theory and Its Applications, IEICE 1(1), 37–68 (2010)
Samaria, F., Harter, A.: Parameterisation of a stochastic model for human face identification. In: Proceedings of 2nd IEEE Workshop on Applications of Computer Vision (1994)
Tucker, L.R.: Some mathematical notes on three-mode factor analysis. Psychometrika 31(3), 279–311 (1966)
Yokota, T., Cichocki, A., Yamashita, Y.: Linked PARAFAC/CP tensor decomposition and its fast implementation for multi-block tensor analysis. In: Huang, T., Zeng, Z., Li, C., Leung, C.S. (eds.) ICONIP 2012, Part III. LNCS, vol. 7665, pp. 84–91. Springer, Heidelberg (2012)
Zhou, G., Cichocki, A., Xie, S.: Common and individual features analysis: beyond canonical correlation analysis. arXiv preprint arXiv:1212.3913 (2012)
Zou, H., Hastie, T., Tibshirani, R.: Sparse principal component analysis. Journal of Computational and Graphical Statistics 15(2), 265–286 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Yokota, T., Cichocki, A. (2014). Linked Tucker2 Decomposition for Flexible Multi-block Data Analysis. In: Loo, C.K., Yap, K.S., Wong, K.W., Beng Jin, A.T., Huang, K. (eds) Neural Information Processing. ICONIP 2014. Lecture Notes in Computer Science, vol 8836. Springer, Cham. https://doi.org/10.1007/978-3-319-12643-2_14
Download citation
DOI: https://doi.org/10.1007/978-3-319-12643-2_14
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-12642-5
Online ISBN: 978-3-319-12643-2
eBook Packages: Computer ScienceComputer Science (R0)