Abstract
In this paper, we present the pseudospectrum for generalized tensor eigenvalues, and a set to locate this pseudospectrum. By the relations between H-eigenvalues (Z-eigenvalues) of tensors and generalized tensor eigenvalues, a pseudospectral localization for H-eigenvalues (Z-eigenvalues, respectively) is given to seek positive definite tensors surrounding a positive definite tensor.
Similar content being viewed by others
References
Bose N, Kamat P (1974) Algorithm for stability test of multidimensional filters. IEEE Trans Acoust Speech Signal Process 22(5):307–314
Bose N, Modarressi A (1976) General procedure for multivariable polynomial positivity test with control applications. IEEE Trans Auto Control 21(5):696–701
Brauer A et al (1947) Limits for the characteristic roots of a matrix. II. Duke Math J 14(1):21–26
Brualdi RA (1982) Matrices eigenvalues, and directed graphs. Linear Multilinear Algebra 11(2):143–165
Bu C, Wei Y, Sun L, Zhou J (2015) Brualdi-type eigenvalue inclusion sets of tensors. Linear Algebra Appl 480:168–175
Bu C, Jin X, Li H, Deng C (2017) Brauer-type eigenvalue inclusion sets and the spectral radius of tensors. Linear Algebra Appl 512:234–248
Chang KC, Pearson K, Zhang T (2008) Perron-Frobenius theorem for nonnegative tensors. Commun Math Sci 6(2):507–520
Chang KC, Pearson K, Zhang T (2009) On eigenvalue problems of real symmetric tensors. J Math Anal Appl 350(1):416–422
Che M, Li G, Qi L, Wei Y (2017a) Pseudo-spectra theory of tensors and tensor polynomial eigenvalue problems. Linear Algebra Appl 533:536–572
Che M, Qi L, Wei Y (2017b) Iterative algorithms for computing us-and u-eigenpairs of complex tensors. J Comput Appl Math 317:547–564
Chen H, Qi L (2015) Positive definiteness and semi-definiteness of even order symmetric cauchy tensors. J Ind Manag Optim 11(4):1263–1274
Chen Y, Dai Y, Han D, Sun W (2013) Positive semidefinite generalized diffusion tensor imaging via quadratic semidefinite programming. SIAM J Imaging Sci 6(3):1531–1552
Ching WK, Ng MK, Ching W (2006) Markov chains: models, algorithms and applications (international series in operations research & management science). Springer, Berlin
Cui CF, Dai YH, Nie J (2014) All real eigenvalues of symmetric tensors. SIAM J Matrix Anal Appl 35(4):1582–1601
De Silva V, Lim LH (2008) Tensor rank and the ill-posedness of the best low-rank approximation problem. SIAM J Matrix Anal Appl 30(3):1084–1127
Deng C, Li H, Bu C (2018) Brauer-type eigenvalue inclusion sets of stochastic/irreducible tensors and positive definiteness of tensors. Linear Algebra Appl 556:55–69
Ding W, Wei Y (2015) Generalized tensor eigenvalue problems. SIAM J Matrix Anal Appl 36(3):1073–1099
Ding W, Qi L, Wei Y (2013) M-tensors and nonsingular M-tensors. Linear Algebra Appl 439(10):3264–3278
Ding W, Hou Z, Wei Y (2016) Tensor logarithmic norm and its applications. Numer Linear Algebra 23(6):989–1006
Friedland S, Mehrmann V, Miedlar A, Nkengla M (2011) Fast low rank approximations of matrices and tensors. Electron J Linear Algebra 22(1):67
Fu M (1998) Comments on “a procedure for the positive definiteness of forms of even order”. IEEE Trans Auto Control 43(10):1430
Gadenz R, Li C (1964) On positive definiteness of quartic forms of two variables. IEEE Trans Auto Control 9(2):187–188
Gershgorin SA (1931) Uber die abgrenzung der eigenwerte einer matrix. Izv Akad Nauk SSSR Ser Mat 6:749–754
Golub G, Loan CV (2013) Matrix computations (4th edn), Johns Hopkins studies in the mathematical sciences. Johns Hopkins University Press, Baltimore
Hasan MA, Hasan AA (1996) A procedure for the positive definiteness of forms of even order. IEEE Trans Auto Control 41(4):615–617
Hillar CJ, Lim LH (2013) Most tensor problems are np-hard. J ACM 60(6):45
Hu S, Qi L (2012) Algebraic connectivity of an even uniform hypergraph. J Comb Optim 24(4):564–579
Hu S, Qi L (2014) The eigenvectors associated with the zero eigenvalues of the Laplacian and signless Laplacian tensors of a uniform hypergraph. Discrete Appl Math 169:140–151
Hu S, Huang Z, Ni H, Qi L (2012) Positive definiteness of diffusion kurtosis imaging. Inverse Probl Imaging 6:57–75
Hu S, Qi L, Shao JY (2013) Cored hypergraphs, power hypergraphs and their Laplacian h-eigenvalues. Linear Algebra Its Appl 439(10):2980–2998
Hu S, Qi L, Zhang G (2016) Computing the geometric measure of entanglement of multipartite pure states by means of non-negative tensors. Phys Rev A 93(1):012304
Hua B, Ni GY, Zhang MS (2017) Computing geometric measure of entanglement for symmetric pure states via the Jacobian sdp relaxation technique. J Oper Res Soc China 5(1):111–121
Kannan MR, Shaked-Monderer N, Berman A (2015) Some properties of strong h-tensors and general h-tensors. Linear Algebra Appl 476:42–55
Kolda TG, Bader BW (2009) Tensor decompositions and applications. SIAM Rev 51(3):455–500
Kolda TG, Mayo JR (2011) Shifted power method for computing tensor eigenpairs. SIAM J Matrix Anal Appl 32(4):1095–1124
Kolda TG, Mayo JR (2014) An adaptive shifted power method for computing generalized tensor eigenpairs. SIAM J Matrix Anal Appl 35(4):1563–1581
Kostić V, Cvetković L, Varga RS (2009) Geršgorin-type localizations of generalized eigenvalues. Numer Linear Algebra 16(11–12):883–898
Kostić V, Cvetković L, Cvetković DL (2016) Pseudospectra localizations and their applications. Numer Linear Algebra 23(2):356–372
Ku W (1965) Explicit criterion for the positive definiteness of a general quartic form. IEEE Trans Auto Control 10(3):372–373
Li C, Li Y (2015) Double b-tensors and quasi-double B-tensors. Linear Algebra Appl 466:343–356
Li C, Li Y (2016) An eigenvalue localization set for tensors with applications to determine the positive (semi-) definiteness of tensors. Linear Multilinear A 64(4):587–601
Li C, Li Y, Kong X (2014a) New eigenvalue inclusion sets for tensors. Numer Linear Algebra 21(1):39–50
Li C, Wang F, Zhao J, Zhu Y, Li Y (2014b) Criterions for the positive definiteness of real supersymmetric tensors. J Comput Appl Math 255:1–14
Li C, Qi L, Li Y (2015) MB-tensors and MB\(_0\)-tensors. Linear Algebra Appl 484:141–153
Li C, Jiao A, Li Y (2016) An S-type eigenvalue localization set for tensors. Linear Algebra Appl 493:469–483
Li W, Ng MK (2014) On the limiting probability distribution of a transition probability tensor. Linear Multilinear A 62(3):362–385
Lim LH (2005) Singular values and eigenvalues of tensors: a variational approach. In: 1st IEEE international workshop on computational advances in multi-sensor adaptive processing, 2005. IEEE, pp 129–132
Lim LH (2008) Spectrum and pseudospectrum of a tensor. University of California, Berkeley
Liu L, Wu W (2006) Dynamical system for computing largest generalized eigenvalue. In: International symposium on neural networks. Springer, pp 399–404
Liu Y, Zhou G, Ibrahim NF (2010) An always convergent algorithm for the largest eigenvalue of an irreducible nonnegative tensor. J Comput Appl Math 235(1):286–292
Ni G, Bai M (2016) Spherical optimization with complex variablesfor computing us-eigenpairs. Comput Optim Appl 65(3):799–820
Ni G, Qi L, Bai M (2014) Geometric measure of entanglement and u-eigenvalues of tensors. SIAM J Matrix Anal Appl 35(1):73–87
Ni Q, Lq Qi, Wang F (2008) An eigenvalue method for the positive definiteness identification problem. IEEE Trans Autom Control 53(5):1096–1107
Nie J, Wang L (2014) Semidefinite relaxations for best rank-1 tensor approximations. SIAM J Matrix Anal Appl 35(3):1155–1179
Qi L (2005) Eigenvalues of a real supersymmetric tensor. J Symb Comput 40(6):1302–1324
Qi L (2011) The best rank-one approximation ratio of a tensor space. SIAM J Matrix Anal Appl 32(2):430–442
Qi L, Luo Z (2017) Tensor analysis: spectral theory and special tensors, vol 151. SIAM, Philadelphia
Qi L, Song Y (2014) An even order symmetric b tensor is positive definite. Linear Algebra Appl 457:303–312
Qi L, Wang Y, Wu EX (2008) D-eigenvalues of diffusion kurtosis tensors. J Comput Appl Math 221(1):150–157
Qi L, Yu G, Wu EX (2010a) Higher order positive semidefinite diffusion tensor imaging. SIAM J Imaging Sci 3(3):416–433
Qi L, Yu G, Wu EX (2010b) Higher order positive semidefinite diffusion tensor imaging. SIAM J Imaging Sci 3(3):416–433
Sang C (2019) A new Brauer-type Z-eigenvalue inclusion set for tensors. Numer Algorithms 80(3):781–794
Song Y, Qi L (2014) Infinite and finite dimensional Hilbert tensors. Linear Algebra Appl 451:1–14
Song Y, Qi L (2015) Properties of some classes of structured tensors. J Optim Theory Appl 165(3):854–873
Sun L, Ji S, Ye J (2008) Hypergraph spectral learning for multi-label classification. In: Proceedings of the 14th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, pp 668–676
Varga RS (2010) Geršgorin and his circles, vol 36. Springer Science & Business Media, Berlin
Wang F, Qi L (2005) Comments on “explicit criterion for the positive definiteness of a general quartic form”. IEEE Trans Autom Control 50(3):416–418
Wang G, Zhou G, Caccetta L (2017) Z-eigenvalue inclusion theorems for tensors. Discrete Cont Dyn-B 22:187–198
Wang X, Navasca C (2018) Low-rank approximation of tensors via sparse optimization. Numer Linear Algebra 25(2):e2136
Wang Y, Wang G (2017) Two s-type z-eigenvalue inclusion sets for tensors. J Inequal Appl 2017(1):152
Wei TC, Goldbart PM (2003) Geometric measure of entanglement and applications to bipartite and multipartite quantum states. Phys Rev A 68(4):042307
Wei Y, Ding W (2016) Theory and computation of tensors: multi-dimensional arrays. Academic Press, Cambridge
Yan S, Xu D, Zhang B, Zhang HJ, Yang Q, Lin S (2007) Graph embedding and extensions: a general framework for dimensionality reduction. IEEE Trans Pattern Anal Mach Intell 29(1):40–51
Zhang L, Qi L, Zhou G (2014) M-tensors and some applications. SIAM J Matrix Anal Appl 35(2):437–452
Acknowledgements
We would like to thank Dr. Maolin Che and Dr. Xuezhong Wang for their useful discussions on this topic. Chaoqian Li is supported partly by the Shanghai Key Laboratory of Contemporary Applied Mathematics under Grant KBH1411209; the National Natural Science Foundation of China under Grant 11601473; the Applied Basic Research Programs of Science and Technology Department of Yunnan Province under Grant 2018FB001; Program for Excellent Young Talents in Yunnan University; Outstanding Youth Cultivation Project for Yunnan Province under Grant 2018YDJQ021; Yunnan Provincial Ten Thousands Plan Young Top Talents. Qilong Liu is supported by Doctoral Scientific Research Foundation of Guizhou Normal University in 2017 under Grant GZNUD [2017]26. Yimin Wei is supported by the National Natural Science Foundation of China under Grant 11771099 and Innovation Program of Shanghai Municipal Education Commission.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Jinyun Yuan.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
We shall show the relations of US-eigenvalues and generalized tensor eigenvalues. Let \( {\mathcal {A}} \in {\mathbb {C}}^{[m,n]}\) be a complex symmetric tensor, and let \( \widehat{{\mathcal {A}}} \in {\mathbb {C}}^{[m,2n]}\) with
and \( \widehat{\mathbf{x }}= \left( \mathbf{x }^\top ,(\mathbf{x }^*)^\top \right) ^\top =({\widehat{x}}_1,\ldots , {\widehat{x}}_{2n})^\top \). When m is even, let
be a tensor such that
for an arbitrary vector \(\mathbf{y }\in {\mathbb {C}}^{2n}\). Since
we rewrite the definition of US-eigenvalues into
This implies that if \(\lambda \) is an US-eigenvalue of an even complex symmetric tensor \({\mathcal {A}}\), then \(\lambda \in \sigma ( \widehat{{\mathcal {A}}}, {\mathcal {J}})\), also see Page 1074-1075 of Ding and Wei (2015).
The \( \epsilon \)-pseudo-spectrum for US-eigenvalues of a complex symmetric tensor \({\mathcal {A}}\in {\mathbb {C}}^{[m,n]}\) is defined as follows:
Clearly, \(\varLambda _0^{US}({\mathcal {A}},|| \cdot ||)\) can be denoted the spectrum for US-eigenvalues of a complex symmetric tensor \({\mathcal {A}}\).
Based the relations between US-eigenvalues with generalized tensor eigenvalues, we give the following set to locate \( \epsilon \)-pseudo-spectrum for US-eigenvalues relative to the tensor infinity norm: \(\varLambda _\epsilon ^{US}({\mathcal {A}},|| \cdot ||_\infty )\).
Theorem 3
Let \({\mathcal {A}}=(a_{i_1\cdots i_m}) \in {\mathbb {C}}^{[m,n]}\) be symmetric with an even m, and \(\epsilon \ge 0\). Then
where
Proof
Let \(\lambda \in \varLambda _\epsilon ^{US}({\mathcal {A}},|| \cdot ||_\infty )\). Then there is a symmetric tensor \({\mathcal {E}}=({\mathcal {E}}_{i_1\cdots i_m})\in {\mathbb {C}}^{[m,n]}\) and a nonzero vector \(\mathbf{x }=(x_1,\ldots ,x_n)^\top \in {\mathbb {C}}^{n} \) such that
By (14), we have
where \(\widehat{{\mathcal {A}}+{\mathcal {E}} } =({\widehat{a}}_{i_1\cdots i_m})\in {\mathbb {C}}^{[m,2n]}\) with
and \({\mathcal {J}}\) is defined as in (13). Furthermore, by Corollary 1 it follows that
Then there is an index \(i_0\in \{1,2,\ldots , 2n\}\) such that
For the tensor \( {\mathcal {J}}\), it is not difficult to see that
and that for any \(i_k\in \{1,2,\ldots ,n\}\) and any \( j_k \in \{n+1,n+2,\ldots ,2n\}, k=1,\ldots ,m\),
On the other hand, for the tensor \(\widehat{{\mathcal {A}}+{\mathcal {E}} } =({\widehat{a}}_{i_1\cdots i_m})\), it is easy to see that for each \(i\in \{1,2,\ldots ,n\}\),
and that for each \(i\in \{n+1,n+2,\ldots ,2n\}\),
Combining (15), (16), (17), (18), (19) and Remark 2, we have that if \(i_0\in \{1,2,\ldots ,n\}\), then
and that if \(i_0\in \{n+1,n+2,\ldots ,2n\}\), then
We now consider the equivalent form of
Taking \(\mathbf{y }=\mathbf{e }=(1,1,\ldots ,1)^\top \in {\mathbb {R}}^{2n} \), then by (13) we have
whenever \(i=1,2\ldots ,2n\). Hence,
Furthermore, note that
and
Then (20), (21), and (22) together give
where
Using the triangle inequality yields
Since \(R_{i_0}({\mathcal {E}})\le || {\mathcal {E}}||_\infty \le \epsilon \), we have
and thus \( \lambda \in \varGamma ^{US}_{j_0}({\mathcal {A}},\epsilon )\in \varGamma ^{US}({\mathcal {A}},\epsilon )\). The conclusion follows. \(\square \)
Rights and permissions
About this article
Cite this article
Li, C., Liu, Q. & Wei, Y. Pseudospectra localizations for generalized tensor eigenvalues to seek more positive definite tensors. Comp. Appl. Math. 38, 183 (2019). https://doi.org/10.1007/s40314-019-0958-6
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s40314-019-0958-6
Keywords
- Pseudospectral localization
- Generalized tensor eigenvalues
- H-eigenvalues
- Z-eigenvalues
- Positive definiteness