Calculation of Fourth-Order Tensor Product Expansion by Power Method and Comparison of it with Higher-Order Singular Value Decomposition

Abstract:

Higher-order singular value decomposition (HOSVD) is known as an effective technique to reduce the dimension of multidimensional data. We have proposed a method to perform third-order tensor product expansion (3OTPE) by using the power method for the same purpose as HOSVD, and showed that our method had a better accuracy property than HOSVD, and furthermore, required fewer computation time than that. Since our method could not be applied to the tensors of fourth-order (or more) in spite of having those useful properties, we extend our algorithm of 3OTPE calculation to forth-order tensors in this paper. The results of newly developed method are compared to those obtained by HOSVD. We show that the results follow the same trend as the case of 3OTPE.

• Periodical: Applied Mechanics and Materials (Volumes 444-445)
• Pages: 703-711
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.444-445.703
• Online since: October 2013
• Authors: Akio Ishida, Takumi Noda, Jun Murakami, Naoki Yamamoto, Chiharu Okuma
• Keywords: Fourth-Order Tensor Product Expansion, Higher-Order Singular Value Decomposition, Multidimensional Data Processing, Power Method

References

[1] D. Muti, S. Bourennane, Multidimensional filtering on a tensor approach, Sig. Proc. 85 (2005) 2338-2353.

DOI: 10.1016/j.sigpro.2004.11.029

[2] Y.M. Lui, Human gesture recognition on product manifolds, J. Mac. Learn. Res. 13 (2012) 3297-3321.

[3] G. Ricci, M.D. Gemmis, G Semeraro, Matrix and tensor factorization techniques applied to recommender systems: a survey, Int. J. Comp. Inf. Technol. 1 (2012) 94-98.

[4] N. Yamamoto, J. Murakami, C. Okuma, Y. Shigeto, S. Saito, T. Izumi, N. Hayashida, Application of multi-dimensional principal component analysis to medical data, Int. J. Eng. Phys. Sci. 6 (2012) 260-266.

[5] L. D. Lathauwer, B. D. Moor, J. Vandewalle, A multilinear singular value decomposition, SIAM J. Mat. Anal. Appl. 21 (2000) 1253-1278.

DOI: 10.1137/s0895479896305696

[6] J. Murakami, N. Yamamoto, Y. Tadokoro, High-speed computation of 3d tensor product expansion by the power method, Electron. Commun. Jpn. 85 (2002) 63-72.

DOI: 10.1002/ecjc.1108

[7] C. Okuma, J. Murakami, N. Yamamoto, An improved algorithm for calculation of the third-order orthogonal tensor product expansion by using singular value decomposition, Int. J. Electron. Commun. Comput. Eng. 2 (2010) 11-20.

[8] C. Okuma, J. Murakami, N. Yamamoto, Comparison between higher-order SVD and third-order orthogonal tensor product expansion, Int. J. Electron. Commun. Comput. Eng. 1 (2009) 131-137.

[9] T. Saitoh, T. Komatsu, H. Harashima, H. Miyakawa, Still picture coding by multi-dimensional outer product expansion (in Japanese), Trans. Inst. Electron. Inf. Commun. Eng. 68-B (1985) 547-548.

[10] G. Strang, Linear Algebra and its Applications, Academic Press, New York, (1976).

[11] Scilab, http: /www. scilab. org.