Advertisement

# Synchronized Recovery Method for Multi-Rank Symmetric Tensor Decomposition

Conference paper
Part of the Mathematics and Visualization book series (MATHVISUAL)

## Abstract

Symmetric tensor decomposition is of great importance in applications. In this paper, we design a synchronized multi-rank symmetric Tensor Decomposition alternating minimization method. In this algorithm, we start from a careful initialization for the non-convex symmetric tensor decomposition and then perform an alternating minimization algorithm. Our contributions are as follows: (1) Our method is synchronized and there is no need for a greedy algorithm to get the multi-rank tensor decomposition . (2) Initialization is an important part in our proposed method. With a careful initialization, our proposed algorithm can converge to the global minimizer of the non-convex objective function. (3) The designed alternating minimization algorithm can give a highly accurate result. In numerical results, our proposed algorithm is much better than the simple gradient descent method from the same initialization. Moreover, our results show that with eigenvectors of random projection as initialization, we can quickly get the global solution by using simple alternating minimization algorithm, though finding the global minimum of this non-convex minimization problem is NP-hard .

## References

1. 1.
A. Anandkumar, R. Ge, D. Hsu, S.M. Kakade, M. Telgarsky. Tensor decompositions for learning latent variable models. J. Mach. Learn. Res. 15(1), 2773–2832 (2014)
2. 2.
K. Batselier, N. Wong, Symmetric tensor decomposition by an iterative eigen decomposition algorithm. J. Comput. Appl. Math. 308, 69–82 (2016)
3. 3.
J. Brachat, P. Comon, B. Mourrain, E. Tsigaridas, Symmetric tensor decomposition, in Signal Processing Conference, 2009 17th European (IEEE, New York, 2009), pp. 525–529Google Scholar
4. 4.
J.-F. Cardoso, Blind signal separation: statistical principles. Proc. IEEE 86(10), 2009–2025 (1998)
5. 5.
P. Chevalier, Optimal separation of independent narrow-band sources: concept and performance. Signal Process. 73(1), 27–47 (1999)
6. 6.
A. Cichocki, S. Amari, Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications, vol. 1 (Wiley, New York, 2002)
7. 7.
P. Comon, M. Rajih, Blind identification of under-determined mixtures based on the characteristic function. Signal Process. 86(9), 2271–2281 (2006)
8. 8.
P. Comon, G. Golub, L.-H. Lim, B. Mourrain, Symmetric tensors and symmetric tensor rank. SIAM J. Matrix Anal. Appl. 30(3), 1254–1279 (2008)
9. 9.
L. De Lathauwer, J. Castaing, Tensor-based techniques for the blind separation of DS-CDMA signals. Signal Process. 87(2), 322–336 (2007)
10. 10.
L. De Lathauwer, B. De Moor, J. Vandewalle, A multilinear singular value decomposition. SIAM J. Matrix Anal. Appl. 21(4), 1253–1278 (2000)
11. 11.
J. Deng, H. Liu, K. Batselier, Y.-K. Kwok, N. Wong, STORM: a nonlinear model order reduction method via symmetric tensor decomposition, in 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC) (IEEE, New York, 2016), pp. 557–562Google Scholar
12. 12.
D.L. Donoho, X. Huo, Uncertainty principles and ideal atomic decomposition. IEEE Trans. Inf. Theory 47(7), 2845–2862 (2001)
13. 13.
A.T. Erdogan, On the convergence of ICA algorithms with symmetric orthogonalization. IEEE Trans. Signal Process. 57(6), 2209–2221 (2009)
14. 14.
A. Ferreol, P. Chevalier, On the behavior of current second and higher order blind source separation methods for cyclostationary sources. IEEE Trans. Signal Process. 48(6), 1712–1725 (2000)
15. 15.
B. Jiang, S. Ma, S. Zhang. Tensor principal component analysis via convex optimization. Math. Programm. 150(2), 423–457 (2015)
16. 16.
E. Kofidis, P.A. Regalia, On the best rank-1 approximation of higher-order supersymmetric tensors. SIAM J. Matrix Anal. Appl. 23(3), 863–884 (2002)
17. 17.
T.G. Kolda, Symmetric orthogonal tensor decomposition is trivial (2015). arXiv preprint arXiv:1503.01375Google Scholar
18. 18.
T.G. Kolda, B.W. Bader, Tensor decompositions and applications. SIAM Rev. 51(3), 455–500 (2009)
19. 19.
T.G. Kolda, J.R. Mayo, Shifted power method for computing tensor eigenpairs. SIAM J. Matrix Anal. Appl. 32(4), 1095–1124 (2011)
20. 20.
T.G. Kolda, J.R. Mayo, An adaptive shifted power method for computing generalized tensor eigenpairs. SIAM J. Matrix Anal. Appl. 35(4), 1563–1581 (2014)
21. 21.
J.B. Kruskal, Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics. Linear Algebra Appl. 18(2), 95–138 (1977)
22. 22.
L.-H. Lim, Singular values and eigenvalues of tensors: a variational approach (2006). arXiv preprint math/0607648Google Scholar
23. 23.
L. Qi, Eigenvalues of a real supersymmetric tensor. J. Symb. Comput. 40(6), 1302–1324 (2005)
24. 24.
L. Qi, F. Wang, Y. Wang, Z-eigenvalue methods for a global polynomial optimization problem. Math. Programm. 118(2), 301–316 (2009)
25. 25.
P.A. Regalia, E. Kofidis, Monotonic convergence of fixed-point algorithms for ICA. IEEE Trans. Neural Netw. 14(4), 943–949 (2003)
26. 26.
N.D. Sidiropoulos, R. Bro, G.B. Giannakis, Parallel factor analysis in sensor array processing. IEEE Trans. Signal Process. 48(8), 2377–2388 (2000)
27. 27.
N.D. Sidiropoulos, G.B. Giannakis, R. Bro. Blind PARAFAC receivers for DS-CDMA systems. IEEE Trans. Signal Process. 48(3), 810–823 (2000)
28. 28.
L.I. Smith, A tutorial on principal components analysis. Cornell University, USA, 51:52 (2002)Google Scholar
29. 29.
A. Swami, G. Giannakis, S. Shamsunder, Multichannel ARMA processes. IEEE Trans. Signal Process. 42(4), 898–913 (1994)
30. 30.
A.-J. Van Der Veen, A. Paulraj, An analytical constant modulus algorithm. IEEE Trans. Signal Process. 44(5), 1136–1155 (1996)

## Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

## Authors and Affiliations

1. 1.School of ScienceHarbin Institute of Technology (Shenzhen)ShenzhenChina
2. 2.Department of MathematicsThe Hong Kong University of Science and TechnologyClear Water Bay, KowloonHong Kong