Sufficient Conditions to Ensure Uniqueness of Best Rank One Approximation

Size: px
Start display at page:

Download "Sufficient Conditions to Ensure Uniqueness of Best Rank One Approximation"

Transcription

1 Sufficient Conditions to Ensure Uniqueness of Best Rank One Approximation Yang Qi Joint work with Pierre Comon and Lek-Heng Lim Supported by ERC grant # Decoda August 3, 2015 Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

2 Overview 1 Motivation 2 Normalized Singular Vector Tuples and Eigenvectors 3 Sufficient Conditions for Uniqueness of Best Rank One Symmetric Approximations 4 Sufficient Conditions for Uniqueness of Best Rank One Approximation Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

3 Overview 1 Motivation 2 Normalized Singular Vector Tuples and Eigenvectors 3 Sufficient Conditions for Uniqueness of Best Rank One Symmetric Approximations 4 Sufficient Conditions for Uniqueness of Best Rank One Approximation Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

4 Motivation Let V 1,..., V d be vector spaces over a field K(= R, C) of dimensions n 1,..., n d respectively. When K = R, fix a basis for each V i, then V 1 V d is naturally endowed with a basis. Given T V 1 V d, let [T i1...i d ] denote the coordinate of T under this basis. Definition If each T i1...i d Definition 0, T is called a nonnegative tensor. The nonnegative rank, denoted by rank + (T ), of a nonnegative tensor T is the minimal integer r such that T = r u 1,i u d,i i=1 where each vector u j,i is nonnegative. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

5 Three Fundamential Questions For any T V 1 V d over K, in engineering it is important to consider a best rank-r approximation T of T, which is a tensor with rank r satisfying T T = inf T X, rank(x ) r where is the Hilbert-Schmidt (Frobenius) norm. The following are three fundamental questions need to be answered: 1 Does T have a best rank-r approximation? 2 when (under what condition) is T unique? 3 How to find this T? Proposition (Lim-Comon) The set of nonnegative tensors with nonnegative rank r is a closed subset. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

6 Matrix Case For a m n matrix T, let T = UΣV be its singular value decomposition, where U and V are unitary matrices, and Σ = diag{σ 1,..., σ k, 0,..., 0} is a non-negative diagonal m n real matrix with σ 1 σ k. Theorem (Schmidt-Eckart-Young) A best rank-r approximation of T is UΛV, where Λ = diag{σ 1,..., σ r, 0,..., 0}. Consequentially, a best rank-r approximation of T can be obtained from r successive best rank one approximations. Definition (Vannieuwenhoven-Nicaise-Vandebril-Meerbergen) Such a T is said to have a Schmidt-Eckart-Young decomposition. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

7 Tensor Case For a tensor, things are quite different from the matrix case. Theorem (Vannieuwenhoven-Nicaise-Vandebril-Meerbergen) The set of tensors that do not admit a Schmidt-Eckart-Young decomposition is of positive measure. Theorem (Stegeman-Comon) Subtracting a best rank one approximation does not necessarily decrease tensor rank. Theorem (Hillar-Lim) Finding a best rank one approximation of a tensor is a NP hard problem. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

8 Best Rank One Approximations Every tensor has a best rank one approximation. Theorem (Friedland-Ottaviani) Almost every T over R has a unique best rank one approximation. For almost every symmetric T S d V, a best rank one approximation is unique and symmetric. Lemma Given a nonnegative T. Let u 1 u d V 1 V d be a best rank one approximation of T, then u 1,..., u d can be chosen to be nonnegative. Aim of This Talk Sufficient conditions to ensure a unique best rank one approximation. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

9 Overview 1 Motivation 2 Normalized Singular Vector Tuples and Eigenvectors 3 Sufficient Conditions for Uniqueness of Best Rank One Symmetric Approximations 4 Sufficient Conditions for Uniqueness of Best Rank One Approximation Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

10 Critical Points of Distance Let S n i 1 be the unit sphere in V i, then a best rank one approximation of T is a solution of the following optimization problem: Lemma (Lim) min min T α v 1 v v i S n i 1 d (1) α 0 A solution of the optimization problem (1) satisfies T, v 1 v i v d = λv i (2) for all 1 i d and some λ R, where, means contraction and means omission. Remark Equation (2) is equivalent to T p T p X, where p = v1 v d is in the cone X of the Segre variety X = Seg(PV 1 PV d ). Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

11 Singular Vector Tuples and Eigenvectors Definition (Friedland-Ottaviani) For T V 1 V d over C, ([u 1 ],..., [u d ]) PV 1 PV d is called a singular vector tuple if for some λ i C, i = 1,..., d. Definition (Lim, L. Qi) T, u 1 û i u d = λ i u i (3) For a symmetric tensor T S d V, λ is called an eigenvalue, and a unit vector v V is called an eigenvector corresponding to λ if T, v d 1 = λv. (4) Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

12 Number of Critical Points Theorem (Cartwright-Sturmfels) For a (general) symmetric T S d V with dim V = n, T has (d 1)n 1 d 2 (equivalence classes of) eigenpairs counted with multiplicity over C. Theorem (Friedland-Ottaviani) For a generic T V 1 V d, T has c simple singular vector tuples corresponding to nonzero singular values, where c is the coefficient of the monomial d i=1 tn i 1 i in 1 j d s n j j t n j j s j t j, s j = k j t k. Remark (Draisma-Horobeț-Ottaviani-Sturmfels-Thomas) These theorems tell us the Euclidean distance degree (ED degree) of the Veronese variety ν d (P n 1 ) is (d 1)n 1 d 2 and the ED degree of the Segre variety Seg(PV 1 PV d ) is c. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

13 Normalized Singular Pairs and Eigenpairs Definition For T V 1 V d, (λ, u 1,..., u d ) is called a normalized singular pair of T if { T, u 1 û i u d = λu i (5) u i, u i = 1 for i = 1,..., d. λ is called a normalized singular value, and (u 1,..., u d ) is called a normalized singular vector tuple corresponding to λ. Definition For T S d W over C, (λ, u) is called a normalized eigenpair of T if { T, u d 1 = λu u, u = 1 Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

14 Equivalence Relations Definition Two normalized eigenpairs (λ, u) and (µ, v) of T are called equivalent if (λ, u) = (µ, v), or ( 1) d 2 λ = µ and u = v. Definition Two normalized singular pairs (λ, u 1,..., u d ) and (µ, v 1,..., v d ) of T are called equivalent if (λ, u 1,..., u d ) = (µ, v 1,..., v d ), or ( 1) k λ = µ and u j = v j for j {i 1,..., i k } and u l = v l for l / {i 1,..., i k }. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

15 Overview 1 Motivation 2 Normalized Singular Vector Tuples and Eigenvectors 3 Sufficient Conditions for Uniqueness of Best Rank One Symmetric Approximations 4 Sufficient Conditions for Uniqueness of Best Rank One Approximation Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

16 Definition (Hillar-Lim) For T V 1 V d over R, let σ 1 (T ) = max{ T, u 1 u d : u 1 = = u d = 1}, and call σ 1 (T ) the spectral norm of T. σ 1 (T ) is the maximal absolute value of normalized singular values of T. Theorem (Friedland-Stawiska) For a closed semi-algebraic set C R m, let S(C) denote the set of points x R m whose best approximation in C is not unique, then S(C) is a nowhere dense semi-algebraic set and contained in a hypersurface. Proposition (Q.-Comon-Lim) Over R, the subset H σ1 S d V consisting of symmetric tensors which have non-unique equivalence classes of normalized eigenpairs corresponding to their spectral norms is indeed a hypersurface in S d V. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

17 Resultant Theory Given n + 1 homogeneous polynomials F 0,..., F n C[x 0,..., x n ] with positive total degrees d 0,..., d n, let F i = α =d i c i,α x α. For each pair of indices i, α, we introduce a variable u i,α. Given a polynomial P C[u i,α ], let P(F 0,..., F n ) denote the number obtained by replacing each variable u i,α in P with the corresponding coefficient c i,α. Theorem There is a unique polynomial Res Z[u i,α ] which has the following properties: 1 The equations F 0 = = F n = 0 have a nonzero solution over C if and only if Res(F 0,..., F n ) = 0. 2 Res(x d 0 0,..., x dn n ) = 1. 3 Res is irreducible in C[u i,α ]. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

18 Characteristic Polynomial Definition (L. Qi) Over C, for T S d V and d = 2l, let ψ T (λ) be the resultant of the equation T, u d 1 λ u, u l 1 u = 0. For d odd, let ψ T (λ) be the resultant of the equations { T, u d 1 λx d 2 u = 0 x 2. u, u = 0 ψ T (λ) is called the E-characteristic polynomial of T. Definition Let χ(t ) denote the resultant of the equations ψ T (λ) = ψ T (λ) = 0. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

19 Proposition (Cartwright-Sturmfels) A general T S d V has distinct normalized eigenvalues. Consequence χ(t ) is a nonzero polynomial on S d V, which implies χ(t ) = 0 defines a complex hypersurface H consisting of tensors T S d (V R C) which admit multiple equivalence classes of normalized eigenpairs. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

20 Example in S 3 R 2 For T = [T jkl ] S 3 R 2, ψ T (λ) = c 2 λ 6 + c 4 λ 4 + c 6 λ 2 + c 8 for some homogeneous polynomials c i of degree i in T jkl. Then χ(t ) is the determinant of some matrix in T jkl. For a generic T, ψ T (λ) = c(λ 2 γ 1 )(λ 2 γ 2 )(λ 2 γ 3 ) for some c C and distinct γ i C, and χ(t ) 0. For T H, ψ T (λ) has multiple roots, χ(t ) = 0. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

21 Example in S 3 R 2 Example Let T by T 111 = T 222 = 1 and other terms T ijk = 0, then χ(t ) = 0, which implies T has multiple eigenpairs. In fact, ψ T (λ) = (λ + 1) 2 (λ 1) 2 (2λ 2 1), there are two eigenvectors (1, 0) and (0, 1) corresponding to the eigenvalue 1, and two eigenvectors ( 1, 0) and (0, 1) corresponding to the eigenvalue 1. This example corresponds to the case that x 3 + y 3 has two best rank one approximations x 3 and y 3, where x and y are two linearly independent vectors. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

22 Conditions for Symmetric Case All real normalized eigenvalues σ 1 σ k of T are roots of ψ T (λ), and in fact for each σ i we have Proposition (Q.-Comon-Lim) 1 Over R, the subset H σi consisting of symmetric tensors which have non-unique equivalence classes of normalized eigenpairs corresponding to σ i is indeed a hypersurface in S d V. 2 These H σi form the components of the real points of H, i.e. H σ1 H σk = real points ofh 3 If χ(t ) 0, T has a unique best rank one approximation which is symmetric. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

23 Overview 1 Motivation 2 Normalized Singular Vector Tuples and Eigenvectors 3 Sufficient Conditions for Uniqueness of Best Rank One Symmetric Approximations 4 Sufficient Conditions for Uniqueness of Best Rank One Approximation Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

24 Singular Characteristic Polynomial Definition Let W 1,..., W d be complex vector spaces, for T W1 W d, u i W i, and α i C, denote by φ T (λ) the resultant of the following homogeneous equations { α i T, u 1 û i u d = λ( j i α j)u i u i, u i = αi 2. (6) By the resultant theory, φ T (λ) vanishes if and only if Equations 6 has a nontrivial solution, i.e. φ T (λ) = 0 if and only if (λ, u 1,..., u d ) is a normalized singular pair of T. φ T (λ) is called the singular characteristic polynomial of T, whose roots λ are normalized singular values of T. Definition Let χ(t ) denote the resultant of the equations φ T (λ) = φ T (λ) = 0. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

25 Proposition (Q.-Comon-Lim) The subset X σ1 V1 V d consisting of tensors which have non-unique best rank one approximations forms an algebraic hypersurface. Proposition (Q.-Comon-Lim) A general T W 1 W d over C has distinct equivalence classes of normalized singular pairs. Therefore for a generic T, φ T (λ) has simple roots, which implies χ(t ) does not vanish identically. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

26 Sufficient Conditions for Uniqueness Since χ(t ) vanishes on X σ1, and does not vanish identically, then χ(t ) = 0 defines a hypersurface in V 1 V d over C. Theorem (Q.-Comon-Lim) The subset M formed by any tensor whose singular characteristic polynomial has multiple roots is a hypersurface, which is defined by χ(t ) = 0. X σ1 forms some components of the real points of M. Any real tensor T with χ(t ) 0 has a unique best rank one approximation. If this T is symmetric, then its unique best rank one approximation is also symmetric. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

27 Thank you very much for your attention! Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

28 S. Banach, Über homogene polynome in (L2 ), Studia Math., 7 (1938), no. 1, pp M. Banagl, The tensor product of function semimodules, Algebra universalis, vol. 70, no. 3, pp , C. Bocci, L. Chiantini, and G. Ottaviani, Refined methods for the identifiability of tensors, Annali di Matematica Pura ed Applicata, May 2013 D. Cartwright, B. Sturmfels, The number of eigenvalues of a tensor, Linear Algebra Appl., no. 438, pp , 2013 K. C. Chang, K. Pearson, and T. Zhang, Perron Frobenius theorem for nonnegative tensors, Commun. Math. Sci., vol. 6, no. 2, pp , L. Chiantini, G. Ottaviani, and N. Vannieuwenhoven, An algorithm for generic and low-rank specific identifiability of complex tensors, SIAM J. matrix Ana. Appl., 2014, to appear. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

29 P. Comon, Tensors: a brief introduction, IEEE Sig. Proc. Magazine, vol. 31, no. 3, pp , May 2014, special issue on BSS. hal D. A. Cox, J. B. Little, and D. O Shea, Using Algebraic Geometry, ser. Graduate Texts in Mathematics, Springer, 2nd ed. 2005, Vol V. De Silva and L.-H. Lim, Tensor rank and the ill-posedness of the best low-rank approximation problem, SIAM Journal on Matrix Analysis Appl., vol. 30, no. 3, pp , S. Friedland, Best rank-1 approximation of real symmetric tensors can be chosen symmetric, Frontiers of Mathematics in China, vol. 8, no. 1, pp , S. Friedland, S. Gaubert, and L. Han, Perron Frobenius theorem for nonnegative multilinear forms and extensions, Linear Algebra Appl., vol. 438, pp , S. Friedland and G. Ottaviani, The number of singular vector tuples and uniqueness of best rank-one approximation of tensors, Mar. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

30 2014, arxiv: , to appear in Foundations of Computational Mathematics. S. Friedland and M. Stawiska, Best approximation on semialgebraic sets and k-border rank approximation of symmetric tensors, Nov. 2013, arxiv: L. D. Garcia, M. Stillman, and B. Sturmfels, Algebraic geometry of Bayesian networks, J. Symbolic Comput., 39 (2005), no. 3 4, pp I. M. Gelfand, M. M. Kapranov, and A. V. Zelevinsky, Discriminants, Resultants, and Multidimensional Determinants, ser. Mathematics: Theory & Applications, Birkhäuser Boston, W. Hackbusch, Tensor Spaces and Numerical Tensor Calculus, ser. Series in Computational Mathematics. Berlin, Heidelberg: Springer, C. J. Hillar and L.-H. Lim, Most tensor problems are NP-hard, Jour. of the ACM, vol. 60, issue 6, no. 45, Nov Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

31 M. I. Jordan, Graphical models, Statistical Science, vol. 19, no. 1, pp , D. Koller and N. Friedman, Probabilistic Graphical Models. MIT Press, J. B. Kruskal, Three-way arrays: Rank and uniqueness of trilinear decompositions, Linear Algebra and Applications, vol. 18, pp , J. M. Landsberg, Tensors: Geometry and Applications, ser. Graduate Studies in Mathematics. AMS publ., 2012, vol A.-M. Li, L. Qi, and B. Zhang, E-characteristic polynomials of tensors, in Commun. Math. Sci., vol. 11, no. 1, pp , L.-H. Lim, Tensors and hypermatrices, Handbook of Linear Algebra, 2nd Ed., CRC Press, Boca Raton, FL, L.-H. Lim, Singular values and eigenvalues of tensors: a variational approach, in IEEE Int. Workshop on Comput. Adv. Multi-Sensor Adapt. Proc., Puerto Vallarta, Mexico, Dec Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

32 L.-H. Lim and P. Comon, Nonnegative approximations of nonnegative tensors, Jour. Chemometrics, vol. 23, pp , Aug. 2009, hal [Online]. Available: L. Oeding and G. Ottaviani, Eigenvectors of tensors and algorithms for Waring decomposition, J. Symbolic Comput., vol. 54, pp. 9 35, 2013 L. Qi, Eigenvalues of a real spersymmetric tensor, J. Symbolic Comput., vol. 40, issue. 6, pp , Dec L. Qi, Eigenvalues and invariants of tensors, J. Math. Anal. Appl., vol. 325, issue. 2, pp , Jan A. Shashua and T. Hazan, Non-negative tensor factorization with applications to statistics and computer vision, in 22nd International Conference on Machine Learning, Bonn, 2005, pp Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

33 N. D. Sidiropoulos and R. Bro, On the uniqueness of multilinear decomposition of N-way arrays, Jour. Chemo., vol. 14, pp , A. Smilde, R. Bro, and P. Geladi, Multi-Way Analysis. Chichester UK: Wiley, A. Stegeman and P. Comon, Subtracting a best rank-1 approximation does not necessarily decrease tensor rank, Linear Algebra Appl., vol. 433, no. 7, pp , Dec. 2010, hal N. Vannieuwenhoven, J. Nicaise, R. Vandebril, and K. Meerbergen, On generic nonexistence of the Schmidt Eckart Young decomposition for complex tensors, SIAM Journal on Matrix Analysis and Applications, vol. 35, no. 3, pp , M. Velasco, Linearization functors on real convex sets, SIAM Journal on Optimization, to appear. Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

34 Y. Yang and Q. Yang, Further Results for Perron-Frobenius Theorem for Nonnegative Tensors, SIAM. J. Matrix Anal. & Appl., vol. 31, issue. 5, pp , P. Zhang, H. Wang, R. Plemmons, and P. Pauca, Tensor methods for hyperspectral data analysis: A space object material identification study, J. Optical Soc. Amer., vol. 25, no. 12, pp , Dec J. Zhou, A. Bhattacharya, A. Herring, and D. Dunson, Bayesian factorizations of big sparse tensors, Jun. 2013, arxiv: Yang Qi (GIPSA-Lab) Uniqueness of Best Rank One Approximation August 3, / 23

The number of singular vector tuples and approximation of symmetric tensors / 13

The number of singular vector tuples and approximation of symmetric tensors / 13 The number of singular vector tuples and approximation of symmetric tensors Shmuel Friedland Univ. Illinois at Chicago Colloquium NYU Courant May 12, 2014 Joint results with Giorgio Ottaviani and Margaret

More information

Eigenvalues of the Adjacency Tensor on Products of Hypergraphs

Eigenvalues of the Adjacency Tensor on Products of Hypergraphs Int. J. Contemp. Math. Sciences, Vol. 8, 201, no. 4, 151-158 HIKARI Ltd, www.m-hikari.com Eigenvalues of the Adjacency Tensor on Products of Hypergraphs Kelly J. Pearson Department of Mathematics and Statistics

More information

Giorgio Ottaviani. Università di Firenze. Tutorial on Tensor rank and tensor decomposition

Giorgio Ottaviani. Università di Firenze. Tutorial on Tensor rank and tensor decomposition Tutorial: A brief survey on tensor rank and tensor decomposition, from a geometric perspective. Workshop Computational nonlinear Algebra (June 2-6, 2014) ICERM, Providence Università di Firenze Tensors

More information

Tensors. Notes by Mateusz Michalek and Bernd Sturmfels for the lecture on June 5, 2018, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra

Tensors. Notes by Mateusz Michalek and Bernd Sturmfels for the lecture on June 5, 2018, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra Tensors Notes by Mateusz Michalek and Bernd Sturmfels for the lecture on June 5, 2018, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra This lecture is divided into two parts. The first part,

More information

THE PERTURBATION BOUND FOR THE SPECTRAL RADIUS OF A NON-NEGATIVE TENSOR

THE PERTURBATION BOUND FOR THE SPECTRAL RADIUS OF A NON-NEGATIVE TENSOR THE PERTURBATION BOUND FOR THE SPECTRAL RADIUS OF A NON-NEGATIVE TENSOR WEN LI AND MICHAEL K. NG Abstract. In this paper, we study the perturbation bound for the spectral radius of an m th - order n-dimensional

More information

Perron-Frobenius theorem for nonnegative multilinear forms and extensions

Perron-Frobenius theorem for nonnegative multilinear forms and extensions Perron-Frobenius theorem for nonnegative multilinear forms and extensions Shmuel Friedland Univ. Illinois at Chicago ensors 18 December, 2010, Hong-Kong Overview 1 Perron-Frobenius theorem for irreducible

More information

Permutation transformations of tensors with an application

Permutation transformations of tensors with an application DOI 10.1186/s40064-016-3720-1 RESEARCH Open Access Permutation transformations of tensors with an application Yao Tang Li *, Zheng Bo Li, Qi Long Liu and Qiong Liu *Correspondence: liyaotang@ynu.edu.cn

More information

Computational Complexity of Tensor Problems

Computational Complexity of Tensor Problems Computational Complexity of Tensor Problems Chris Hillar Redwood Center for Theoretical Neuroscience (UC Berkeley) Joint work with Lek-Heng Lim (with L-H Lim) Most tensor problems are NP-hard, Journal

More information

A Geometric Perspective on the Singular Value Decomposition

A Geometric Perspective on the Singular Value Decomposition arxiv:1503.07054v1 [math.ag] 24 Mar 2015 A Geometric Perspective on the Singular Value Decomposition Giorgio Ottaviani and Raffaella Paoletti dedicated to Emilia Mezzetti Abstract This is an introductory

More information

Spectrum and Pseudospectrum of a Tensor

Spectrum and Pseudospectrum of a Tensor Spectrum and Pseudospectrum of a Tensor Lek-Heng Lim University of California, Berkeley July 11, 2008 L.-H. Lim (Berkeley) Spectrum and Pseudospectrum of a Tensor July 11, 2008 1 / 27 Matrix eigenvalues

More information

Eigenvalues of tensors

Eigenvalues of tensors Eigenvalues of tensors and some very basic spectral hypergraph theory Lek-Heng Lim Matrix Computations and Scientific Computing Seminar April 16, 2008 Lek-Heng Lim (MCSC Seminar) Eigenvalues of tensors

More information

Hyperdeterminants, secant varieties, and tensor approximations

Hyperdeterminants, secant varieties, and tensor approximations Hyperdeterminants, secant varieties, and tensor approximations Lek-Heng Lim University of California, Berkeley April 23, 2008 Joint work with Vin de Silva L.-H. Lim (Berkeley) Tensor approximations April

More information

Strictly nonnegative tensors and nonnegative tensor partition

Strictly nonnegative tensors and nonnegative tensor partition SCIENCE CHINA Mathematics. ARTICLES. January 2012 Vol. 55 No. 1: 1 XX doi: Strictly nonnegative tensors and nonnegative tensor partition Hu Shenglong 1, Huang Zheng-Hai 2,3, & Qi Liqun 4 1 Department of

More information

Oeding (Auburn) tensors of rank 5 December 15, / 24

Oeding (Auburn) tensors of rank 5 December 15, / 24 Oeding (Auburn) 2 2 2 2 2 tensors of rank 5 December 15, 2015 1 / 24 Recall Peter Burgisser s overview lecture (Jan Draisma s SIAM News article). Big Goal: Bound the computational complexity of det n,

More information

Tensors. Lek-Heng Lim. Statistics Department Retreat. October 27, Thanks: NSF DMS and DMS

Tensors. Lek-Heng Lim. Statistics Department Retreat. October 27, Thanks: NSF DMS and DMS Tensors Lek-Heng Lim Statistics Department Retreat October 27, 2012 Thanks: NSF DMS 1209136 and DMS 1057064 L.-H. Lim (Stat Retreat) Tensors October 27, 2012 1 / 20 tensors on one foot a tensor is a multilinear

More information

c 2016 Society for Industrial and Applied Mathematics

c 2016 Society for Industrial and Applied Mathematics SIAM J. MATRIX ANAL. APPL. Vol. 37, No. 4, pp. 1556 1580 c 2016 Society for Industrial and Applied Mathematics SEMIALGEBRAIC GEOMETRY OF NONNEGATIVE TENSOR RANK YANG QI, PIERRE COMON, AND LEK-HENG LIM

More information

Tensors: a geometric view Open lecture November 24, 2014 Simons Institute, Berkeley. Giorgio Ottaviani, Università di Firenze

Tensors: a geometric view Open lecture November 24, 2014 Simons Institute, Berkeley. Giorgio Ottaviani, Università di Firenze Open lecture November 24, 2014 Simons Institute, Berkeley Plan of the talk Introduction to Tensors A few relevant applications Tensors as multidimensional matrices A (complex) a b c tensor is an element

More information

TOPOLOGY OF TENSOR RANKS

TOPOLOGY OF TENSOR RANKS TOPOLOGY OF TENSOR RANKS PIERRE COMON, LEK-HENG LIM, YANG QI, AND KE YE Abstract. We study path-connectedness and homotopy groups of sets of tensors defined by tensor rank, border rank, multilinear rank,

More information

Numerical multilinear algebra

Numerical multilinear algebra Numerical multilinear algebra From matrices to tensors Lek-Heng Lim University of California, Berkeley March 1, 2008 Lek-Heng Lim (UC Berkeley) Numerical multilinear algebra March 1, 2008 1 / 29 Lesson

More information

ON THE LIMITING PROBABILITY DISTRIBUTION OF A TRANSITION PROBABILITY TENSOR

ON THE LIMITING PROBABILITY DISTRIBUTION OF A TRANSITION PROBABILITY TENSOR ON THE LIMITING PROBABILITY DISTRIBUTION OF A TRANSITION PROBABILITY TENSOR WEN LI AND MICHAEL K. NG Abstract. In this paper we propose and develop an iterative method to calculate a limiting probability

More information

R, 1 i 1,i 2,...,i m n.

R, 1 i 1,i 2,...,i m n. SIAM J. MATRIX ANAL. APPL. Vol. 31 No. 3 pp. 1090 1099 c 2009 Society for Industrial and Applied Mathematics FINDING THE LARGEST EIGENVALUE OF A NONNEGATIVE TENSOR MICHAEL NG LIQUN QI AND GUANGLU ZHOU

More information

Eigenvectors of Tensors and Waring Decomposition

Eigenvectors of Tensors and Waring Decomposition Eigenvectors of Tensors and Waring Decomposition Luke Oeding University of California Berkeley Oeding, Ottaviani (UCB, Firenze) Giorgio Ottaviani Universita degli Studi di Firenze Waring Decomposition

More information

Approximation algorithms for nonnegative polynomial optimization problems over unit spheres

Approximation algorithms for nonnegative polynomial optimization problems over unit spheres Front. Math. China 2017, 12(6): 1409 1426 https://doi.org/10.1007/s11464-017-0644-1 Approximation algorithms for nonnegative polynomial optimization problems over unit spheres Xinzhen ZHANG 1, Guanglu

More information

Finding the Maximum Eigenvalue of Essentially Nonnegative Symmetric Tensors via Sum of Squares Programming

Finding the Maximum Eigenvalue of Essentially Nonnegative Symmetric Tensors via Sum of Squares Programming Finding the Maximum Eigenvalue of Essentially Nonnegative Symmetric Tensors via Sum of Squares Programming Shenglong Hu Guoyin Li Liqun Qi Yisheng Song September 16, 2012 Abstract Finding the maximum eigenvalue

More information

arxiv: v1 [math.ra] 11 Aug 2014

arxiv: v1 [math.ra] 11 Aug 2014 Double B-tensors and quasi-double B-tensors Chaoqian Li, Yaotang Li arxiv:1408.2299v1 [math.ra] 11 Aug 2014 a School of Mathematics and Statistics, Yunnan University, Kunming, Yunnan, P. R. China 650091

More information

Multilinear Algebra in Data Analysis:

Multilinear Algebra in Data Analysis: Multilinear Algebra in Data Analysis: tensors, symmetric tensors, nonnegative tensors Lek-Heng Lim Stanford University Workshop on Algorithms for Modern Massive Datasets Stanford, CA June 21 24, 2006 Thanks:

More information

Newcriteria for H-tensors and an application

Newcriteria for H-tensors and an application Wang and Sun Journal of Inequalities and Applications (016) 016:96 DOI 10.1186/s13660-016-1041-0 R E S E A R C H Open Access Newcriteria for H-tensors and an application Feng Wang * and De-shu Sun * Correspondence:

More information

Eigenvalues and invariants of tensors

Eigenvalues and invariants of tensors J. Math. Anal. Appl. 325 (2007) 1363 1377 www.elsevier.com/locate/jmaa Eigenvalues and invariants of tensors Liqun Q Department of Applied Mathematics, The Hong Kong Polytechnic University, Kowloon, Hong

More information

Spectra of Random Symmetric Hypermatrices and Hypergraphs

Spectra of Random Symmetric Hypermatrices and Hypergraphs Spectra of Random Symmetric Hypermatrices and Hypergraphs Joshua Cooper August 21, 2015 Abstract We discuss progress on the problem of asymptotically describing the complex homogeneous adjacency eigenvalues

More information

Is Every Secant Variety of a Segre Product Arithmetically Cohen Macaulay? Oeding (Auburn) acm Secants March 6, / 23

Is Every Secant Variety of a Segre Product Arithmetically Cohen Macaulay? Oeding (Auburn) acm Secants March 6, / 23 Is Every Secant Variety of a Segre Product Arithmetically Cohen Macaulay? Luke Oeding Auburn University Oeding (Auburn) acm Secants March 6, 2016 1 / 23 Secant varieties and tensors Let V 1,..., V d, be

More information

EXISTENCE OF NON-SUBNORMAL POLYNOMIALLY HYPONORMAL OPERATORS

EXISTENCE OF NON-SUBNORMAL POLYNOMIALLY HYPONORMAL OPERATORS BULLETIN (New Series) OF THE AMERICAN MATHEMATICAL SOCIETY Volume 25, Number 2, October 1991 EXISTENCE OF NON-SUBNORMAL POLYNOMIALLY HYPONORMAL OPERATORS RAUL E. CURTO AND MIHAI PUTINAR INTRODUCTION In

More information

Ranks of Real Symmetric Tensors

Ranks of Real Symmetric Tensors Ranks of Real Symmetric Tensors Greg Blekherman SIAM AG 2013 Algebraic Geometry of Tensor Decompositions Real Symmetric Tensor Decompositions Let f be a form of degree d in R[x 1,..., x n ]. We would like

More information

A SURVEY ON THE SPECTRAL THEORY OF NONNEGATIVE TENSORS

A SURVEY ON THE SPECTRAL THEORY OF NONNEGATIVE TENSORS A SURVEY ON THE SPECTRAL THEORY OF NONNEGATIVE TENSORS K.C. CHANG, LIQUN QI, AND TAN ZHANG Abstract. This is a survey paper on the recent development of the spectral theory of nonnegative tensors and its

More information

Conditioning and illposedness of tensor approximations

Conditioning and illposedness of tensor approximations Conditioning and illposedness of tensor approximations and why engineers should care Lek-Heng Lim MSRI Summer Graduate Workshop July 7 18, 2008 L.-H. Lim (MSRI SGW) Conditioning and illposedness July 7

More information

Inverse Perron values and connectivity of a uniform hypergraph

Inverse Perron values and connectivity of a uniform hypergraph Inverse Perron values and connectivity of a uniform hypergraph Changjiang Bu College of Automation College of Science Harbin Engineering University Harbin, PR China buchangjiang@hrbeu.edu.cn Jiang Zhou

More information

A new truncation strategy for the higher-order singular value decomposition

A new truncation strategy for the higher-order singular value decomposition A new truncation strategy for the higher-order singular value decomposition Nick Vannieuwenhoven K.U.Leuven, Belgium Workshop on Matrix Equations and Tensor Techniques RWTH Aachen, Germany November 21,

More information

MATRIX COMPLETION AND TENSOR RANK

MATRIX COMPLETION AND TENSOR RANK MATRIX COMPLETION AND TENSOR RANK HARM DERKSEN Abstract. In this paper, we show that the low rank matrix completion problem can be reduced to the problem of finding the rank of a certain tensor. arxiv:1302.2639v2

More information

Globally convergent algorithms for PARAFAC with semi-definite programming

Globally convergent algorithms for PARAFAC with semi-definite programming Globally convergent algorithms for PARAFAC with semi-definite programming Lek-Heng Lim 6th ERCIM Workshop on Matrix Computations and Statistics Copenhagen, Denmark April 1 3, 2005 Thanks: Vin de Silva,

More information

Basic Calculus Review

Basic Calculus Review Basic Calculus Review Lorenzo Rosasco ISML Mod. 2 - Machine Learning Vector Spaces Functionals and Operators (Matrices) Vector Space A vector space is a set V with binary operations +: V V V and : R V

More information

From nonnegative matrices to nonnegative tensors

From nonnegative matrices to nonnegative tensors From nonnegative matrices to nonnegative tensors Shmuel Friedland Univ. Illinois at Chicago 7 October, 2011 Hamilton Institute, National University of Ireland Overview 1 Perron-Frobenius theorem for irreducible

More information

A POLAR, THE CLASS AND PLANE JACOBIAN CONJECTURE

A POLAR, THE CLASS AND PLANE JACOBIAN CONJECTURE Bull. Korean Math. Soc. 47 (2010), No. 1, pp. 211 219 DOI 10.4134/BKMS.2010.47.1.211 A POLAR, THE CLASS AND PLANE JACOBIAN CONJECTURE Dosang Joe Abstract. Let P be a Jacobian polynomial such as deg P =

More information

Most tensor problems are NP hard

Most tensor problems are NP hard Most tensor problems are NP hard (preliminary report) Lek-Heng Lim University of California, Berkeley August 25, 2009 (Joint work with: Chris Hillar, MSRI) L.H. Lim (Berkeley) Most tensor problems are

More information

Uniqueness of Nonnegative Tensor Approximations

Uniqueness of Nonnegative Tensor Approximations Uniqueness of Nonnegative Tensor Approximations Yang Qi, Pierre Comon, Lek-Heng Lim To cite this version: Yang Qi, Pierre Comon, Lek-Heng Lim. Uniqueness of Nonnegative Tensor Approximations. IEEE Transactions

More information

E-Characteristic Polynomials of Tensors

E-Characteristic Polynomials of Tensors E-Characteristic Polynomials of Tensors arxiv:1208.1607v1 [math.sp] 8 Aug 2012 An-Min Li Liqun Qi Bin Zhang August 9, 2012 Abstract In this paper, we show that the coefficients of the E-characteristic

More information

Secant Varieties and Equations of Abo-Wan Hypersurfaces

Secant Varieties and Equations of Abo-Wan Hypersurfaces Secant Varieties and Equations of Abo-Wan Hypersurfaces Luke Oeding Auburn University Oeding (Auburn, NIMS) Secants, Equations and Applications August 9, 214 1 / 34 Secant varieties Suppose X is an algebraic

More information

arxiv: v1 [math.co] 10 Aug 2016

arxiv: v1 [math.co] 10 Aug 2016 POLYTOPES OF STOCHASTIC TENSORS HAIXIA CHANG 1, VEHBI E. PAKSOY 2 AND FUZHEN ZHANG 2 arxiv:1608.03203v1 [math.co] 10 Aug 2016 Abstract. Considering n n n stochastic tensors (a ijk ) (i.e., nonnegative

More information

Eigenvalue analysis of constrained minimization problem for homogeneous polynomial

Eigenvalue analysis of constrained minimization problem for homogeneous polynomial J Glob Optim 2016) 64:563 575 DOI 10.1007/s10898-015-0343-y Eigenvalue analysis of constrained minimization problem for homogeneous polynomial Yisheng Song 1,2 Liqun Qi 2 Received: 31 May 2014 / Accepted:

More information

Pencil-based algorithms for tensor rank decomposition are unstable

Pencil-based algorithms for tensor rank decomposition are unstable Pencil-based algorithms for tensor rank decomposition are unstable September 13, 2018 Nick Vannieuwenhoven (FWO / KU Leuven) Introduction Overview 1 Introduction 2 Sensitivity 3 Pencil-based algorithms

More information

Numerically Testing Generically Reduced Projective Schemes for the Arithmetic Gorenstein Property

Numerically Testing Generically Reduced Projective Schemes for the Arithmetic Gorenstein Property Numerically Testing Generically Reduced Projective Schemes for the Arithmetic Gorenstein Property Noah S. Daleo 1 and Jonathan D. Hauenstein 2 1 Department of Mathematics, North Carolina State University,

More information

STA141C: Big Data & High Performance Statistical Computing

STA141C: Big Data & High Performance Statistical Computing STA141C: Big Data & High Performance Statistical Computing Numerical Linear Algebra Background Cho-Jui Hsieh UC Davis May 15, 2018 Linear Algebra Background Vectors A vector has a direction and a magnitude

More information

Positive systems in the behavioral approach: main issues and recent results

Positive systems in the behavioral approach: main issues and recent results Positive systems in the behavioral approach: main issues and recent results Maria Elena Valcher Dipartimento di Elettronica ed Informatica Università dipadova via Gradenigo 6A 35131 Padova Italy Abstract

More information

Algebraic Varieties. Notes by Mateusz Micha lek for the lecture on April 17, 2018, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra

Algebraic Varieties. Notes by Mateusz Micha lek for the lecture on April 17, 2018, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra Algebraic Varieties Notes by Mateusz Micha lek for the lecture on April 17, 2018, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra Algebraic varieties represent solutions of a system of polynomial

More information

Mathematical foundations - linear algebra

Mathematical foundations - linear algebra Mathematical foundations - linear algebra Andrea Passerini passerini@disi.unitn.it Machine Learning Vector space Definition (over reals) A set X is called a vector space over IR if addition and scalar

More information

Functional Analysis Review

Functional Analysis Review Outline 9.520: Statistical Learning Theory and Applications February 8, 2010 Outline 1 2 3 4 Vector Space Outline A vector space is a set V with binary operations +: V V V and : R V V such that for all

More information

On the Adjacency Spectra of Hypertrees

On the Adjacency Spectra of Hypertrees On the Adjacency Spectra of Hypertrees arxiv:1711.01466v1 [math.sp] 4 Nov 2017 Gregory J. Clark and Joshua N. Cooper Department of Mathematics University of South Carolina November 7, 2017 Abstract We

More information

Subtracting a best rank-1 approximation may increase tensor rank

Subtracting a best rank-1 approximation may increase tensor rank Subtracting a best rank- approximation may increase tensor rank Alwin Stegeman, Pierre Comon To cite this version: Alwin Stegeman, Pierre Comon. Subtracting a best rank- approximation may increase tensor

More information

arxiv: v3 [math.ag] 15 Dec 2016

arxiv: v3 [math.ag] 15 Dec 2016 The number of real eigenvectors of a real polynomial Mauro Maccioni arxiv:606.04737v3 [math.ag] 5 Dec 06 9 November 06 Abstract I investigate on the number t of real eigenvectors of a real symmetric tensor.

More information

arxiv: v3 [math.ra] 22 Aug 2016

arxiv: v3 [math.ra] 22 Aug 2016 arxiv:1601.05351v3 [math.ra] 22 Aug 2016 SEMIALGEBRAIC GEOMETRY OF NONNEGATIVE TENSOR RANK YANG QI, PIERRE COMON, AND LEK-HENG LIM Abstract. We study the semialgebraic structure of D r, the set of nonnegative

More information

GRÖBNER BASES AND POLYNOMIAL EQUATIONS. 1. Introduction and preliminaries on Gróbner bases

GRÖBNER BASES AND POLYNOMIAL EQUATIONS. 1. Introduction and preliminaries on Gróbner bases GRÖBNER BASES AND POLYNOMIAL EQUATIONS J. K. VERMA 1. Introduction and preliminaries on Gróbner bases Let S = k[x 1, x 2,..., x n ] denote a polynomial ring over a field k where x 1, x 2,..., x n are indeterminates.

More information

arxiv: v2 [math-ph] 3 Jun 2017

arxiv: v2 [math-ph] 3 Jun 2017 Elasticity M -tensors and the Strong Ellipticity Condition Weiyang Ding Jinjie Liu Liqun Qi Hong Yan June 6, 2017 arxiv:170509911v2 [math-ph] 3 Jun 2017 Abstract In this paper, we propose a class of tensors

More information

MP 472 Quantum Information and Computation

MP 472 Quantum Information and Computation MP 472 Quantum Information and Computation http://www.thphys.may.ie/staff/jvala/mp472.htm Outline Open quantum systems The density operator ensemble of quantum states general properties the reduced density

More information

Notes on Eigenvalues, Singular Values and QR

Notes on Eigenvalues, Singular Values and QR Notes on Eigenvalues, Singular Values and QR Michael Overton, Numerical Computing, Spring 2017 March 30, 2017 1 Eigenvalues Everyone who has studied linear algebra knows the definition: given a square

More information

POINT VALUES AND NORMALIZATION OF TWO-DIRECTION MULTIWAVELETS AND THEIR DERIVATIVES

POINT VALUES AND NORMALIZATION OF TWO-DIRECTION MULTIWAVELETS AND THEIR DERIVATIVES November 1, 1 POINT VALUES AND NORMALIZATION OF TWO-DIRECTION MULTIWAVELETS AND THEIR DERIVATIVES FRITZ KEINERT AND SOON-GEOL KWON,1 Abstract Two-direction multiscaling functions φ and two-direction multiwavelets

More information

M-tensors and nonsingular M-tensors

M-tensors and nonsingular M-tensors Linear Algebra and its Applications 439 (2013) 3264 3278 Contents lists available at ScienceDirect Linear Algebra and its Applications www.elsevier.com/locate/laa M-tensors and nonsingular M-tensors Weiyang

More information

Spectral directed hypergraph theory via tensors

Spectral directed hypergraph theory via tensors Linear and Multilinear Algebra ISSN: 0308-1087 (Print) 1563-5139 (Online) Journal homepage: http://www.tandfonline.com/loi/glma20 Spectral directed hypergraph theory via tensors Jinshan Xie & Liqun Qi

More information

Max-Planck-Institut für Mathematik in den Naturwissenschaften Leipzig

Max-Planck-Institut für Mathematik in den Naturwissenschaften Leipzig Max-Planck-Institut für Mathematik in den Naturwissenschaften Leipzig Towards a condition number theorem for the tensor rank decomposition by Paul Breiding and Nick Vannieuwenhoven Preprint no.: 3 208

More information

THE BKK ROOT COUNT IN C n. c e x e,

THE BKK ROOT COUNT IN C n. c e x e, MATHEMATICS OF COMPUTATION Volume 65, Number 216 October 1996, Pages 1477 1484 THE BKK ROOT COUNT IN C n T Y LI AND XIAOSHEN WANG Abstract The root count developed by Bernshtein, Kushnirenko and Khovanskii

More information

Characterization of half-radial matrices

Characterization of half-radial matrices Characterization of half-radial matrices Iveta Hnětynková, Petr Tichý Faculty of Mathematics and Physics, Charles University, Sokolovská 83, Prague 8, Czech Republic Abstract Numerical radius r(a) is the

More information

Applied Linear Algebra in Geoscience Using MATLAB

Applied Linear Algebra in Geoscience Using MATLAB Applied Linear Algebra in Geoscience Using MATLAB Contents Getting Started Creating Arrays Mathematical Operations with Arrays Using Script Files and Managing Data Two-Dimensional Plots Programming in

More information

The Singular Value Decomposition

The Singular Value Decomposition The Singular Value Decomposition An Important topic in NLA Radu Tiberiu Trîmbiţaş Babeş-Bolyai University February 23, 2009 Radu Tiberiu Trîmbiţaş ( Babeş-Bolyai University)The Singular Value Decomposition

More information

Geometric Mapping Properties of Semipositive Matrices

Geometric Mapping Properties of Semipositive Matrices Geometric Mapping Properties of Semipositive Matrices M. J. Tsatsomeros Mathematics Department Washington State University Pullman, WA 99164 (tsat@wsu.edu) July 14, 2015 Abstract Semipositive matrices

More information

Tensor networks, TT (Matrix Product States) and Hierarchical Tucker decomposition

Tensor networks, TT (Matrix Product States) and Hierarchical Tucker decomposition Tensor networks, TT (Matrix Product States) and Hierarchical Tucker decomposition R. Schneider (TUB Matheon) John von Neumann Lecture TU Munich, 2012 Setting - Tensors V ν := R n, H d = H := d ν=1 V ν

More information

STA141C: Big Data & High Performance Statistical Computing

STA141C: Big Data & High Performance Statistical Computing STA141C: Big Data & High Performance Statistical Computing Lecture 5: Numerical Linear Algebra Cho-Jui Hsieh UC Davis April 20, 2017 Linear Algebra Background Vectors A vector has a direction and a magnitude

More information

Research Article Localization and Perturbations of Roots to Systems of Polynomial Equations

Research Article Localization and Perturbations of Roots to Systems of Polynomial Equations International Mathematics and Mathematical Sciences Volume 2012, Article ID 653914, 9 pages doi:10.1155/2012/653914 Research Article Localization and Perturbations of Roots to Systems of Polynomial Equations

More information

Z-eigenvalue methods for a global polynomial optimization problem

Z-eigenvalue methods for a global polynomial optimization problem Math. Program., Ser. A (2009) 118:301 316 DOI 10.1007/s10107-007-0193-6 FULL LENGTH PAPER Z-eigenvalue methods for a global polynomial optimization problem Liqun Qi Fei Wang Yiju Wang Received: 6 June

More information

A concise proof of Kruskal s theorem on tensor decomposition

A concise proof of Kruskal s theorem on tensor decomposition A concise proof of Kruskal s theorem on tensor decomposition John A. Rhodes 1 Department of Mathematics and Statistics University of Alaska Fairbanks PO Box 756660 Fairbanks, AK 99775 Abstract A theorem

More information

Some bounds for the spectral radius of the Hadamard product of matrices

Some bounds for the spectral radius of the Hadamard product of matrices Some bounds for the spectral radius of the Hadamard product of matrices Guang-Hui Cheng, Xiao-Yu Cheng, Ting-Zhu Huang, Tin-Yau Tam. June 1, 2004 Abstract Some bounds for the spectral radius of the Hadamard

More information

Real Symmetric Matrices and Semidefinite Programming

Real Symmetric Matrices and Semidefinite Programming Real Symmetric Matrices and Semidefinite Programming Tatsiana Maskalevich Abstract Symmetric real matrices attain an important property stating that all their eigenvalues are real. This gives rise to many

More information

Lecture 8: Linear Algebra Background

Lecture 8: Linear Algebra Background CSE 521: Design and Analysis of Algorithms I Winter 2017 Lecture 8: Linear Algebra Background Lecturer: Shayan Oveis Gharan 2/1/2017 Scribe: Swati Padmanabhan Disclaimer: These notes have not been subjected

More information

MOST TENSOR PROBLEMS ARE NP HARD CHRISTOPHER J. HILLAR AND LEK-HENG LIM

MOST TENSOR PROBLEMS ARE NP HARD CHRISTOPHER J. HILLAR AND LEK-HENG LIM MOST TENSOR PROBLEMS ARE NP HARD CHRISTOPHER J. HILLAR AND LEK-HENG LIM Abstract. The idea that one might extend numerical linear algebra, the collection of matrix computational methods that form the workhorse

More information

Machine Learning. B. Unsupervised Learning B.2 Dimensionality Reduction. Lars Schmidt-Thieme, Nicolas Schilling

Machine Learning. B. Unsupervised Learning B.2 Dimensionality Reduction. Lars Schmidt-Thieme, Nicolas Schilling Machine Learning B. Unsupervised Learning B.2 Dimensionality Reduction Lars Schmidt-Thieme, Nicolas Schilling Information Systems and Machine Learning Lab (ISMLL) Institute for Computer Science University

More information

Ir O D = D = ( ) Section 2.6 Example 1. (Bottom of page 119) dim(v ) = dim(l(v, W )) = dim(v ) dim(f ) = dim(v )

Ir O D = D = ( ) Section 2.6 Example 1. (Bottom of page 119) dim(v ) = dim(l(v, W )) = dim(v ) dim(f ) = dim(v ) Section 3.2 Theorem 3.6. Let A be an m n matrix of rank r. Then r m, r n, and, by means of a finite number of elementary row and column operations, A can be transformed into the matrix ( ) Ir O D = 1 O

More information

AN INVERSE EIGENVALUE PROBLEM AND AN ASSOCIATED APPROXIMATION PROBLEM FOR GENERALIZED K-CENTROHERMITIAN MATRICES

AN INVERSE EIGENVALUE PROBLEM AND AN ASSOCIATED APPROXIMATION PROBLEM FOR GENERALIZED K-CENTROHERMITIAN MATRICES AN INVERSE EIGENVALUE PROBLEM AND AN ASSOCIATED APPROXIMATION PROBLEM FOR GENERALIZED K-CENTROHERMITIAN MATRICES ZHONGYUN LIU AND HEIKE FAßBENDER Abstract: A partially described inverse eigenvalue problem

More information

Numerical Multilinear Algebra: a new beginning?

Numerical Multilinear Algebra: a new beginning? Numerical Multilinear Algebra: a new beginning? Lek-Heng Lim Stanford University Matrix Computations and Scientific Computing Seminar Berkeley, CA October 18, 2006 Thanks: G. Carlsson, S. Lacoste-Julien,

More information

Applications of The Perron-Frobenius Theorem

Applications of The Perron-Frobenius Theorem Applications of The Perron-Frobenius Theorem Nate Iverson The University of Toledo Toledo, Ohio Motivation In a finite discrete linear dynamical system x n+1 = Ax n What are sufficient conditions for x

More information

Tensor decomposition and tensor rank

Tensor decomposition and tensor rank from the point of view of Classical Algebraic Geometry RTG Workshop Tensors and their Geometry in High Dimensions (September 26-29, 2012) UC Berkeley Università di Firenze Content of the three talks Wednesday

More information

An Even Order Symmetric B Tensor is Positive Definite

An Even Order Symmetric B Tensor is Positive Definite An Even Order Symmetric B Tensor is Positive Definite Liqun Qi, Yisheng Song arxiv:1404.0452v4 [math.sp] 14 May 2014 October 17, 2018 Abstract It is easily checkable if a given tensor is a B tensor, or

More information

Throughout these notes we assume V, W are finite dimensional inner product spaces over C.

Throughout these notes we assume V, W are finite dimensional inner product spaces over C. Math 342 - Linear Algebra II Notes Throughout these notes we assume V, W are finite dimensional inner product spaces over C 1 Upper Triangular Representation Proposition: Let T L(V ) There exists an orthonormal

More information

Applied Linear Algebra in Geoscience Using MATLAB

Applied Linear Algebra in Geoscience Using MATLAB Applied Linear Algebra in Geoscience Using MATLAB Contents Getting Started Creating Arrays Mathematical Operations with Arrays Using Script Files and Managing Data Two-Dimensional Plots Programming in

More information

Optimal solutions to non-negative PARAFAC/multilinear NMF always exist

Optimal solutions to non-negative PARAFAC/multilinear NMF always exist Optimal solutions to non-negative PARAFAC/multilinear NMF always exist Lek-Heng Lim Stanford University Workshop on Tensor Decompositions and Applications CIRM, Luminy, France August 29 September 2, 2005

More information

Banach Journal of Mathematical Analysis ISSN: (electronic)

Banach Journal of Mathematical Analysis ISSN: (electronic) Banach J. Math. Anal. 6 (2012), no. 1, 139 146 Banach Journal of Mathematical Analysis ISSN: 1735-8787 (electronic) www.emis.de/journals/bjma/ AN EXTENSION OF KY FAN S DOMINANCE THEOREM RAHIM ALIZADEH

More information

CHARACTERIZATION OF GORENSTEIN STRONGLY KOSZUL HIBI RINGS BY F-INVARIANTS

CHARACTERIZATION OF GORENSTEIN STRONGLY KOSZUL HIBI RINGS BY F-INVARIANTS CHARACTERIZATION OF GORENSTEIN STRONGLY KOSZUL HIBI RINGS BY F-INVARIANTS KAZUNORI MATSUDA Abstract. Hibi rings are a kind of graded toric ring on a finite distributive lattice D = J(P ), where P is a

More information

Wavelets and Linear Algebra

Wavelets and Linear Algebra Wavelets and Linear Algebra 4(1) (2017) 43-51 Wavelets and Linear Algebra Vali-e-Asr University of Rafsanan http://walavruacir Some results on the block numerical range Mostafa Zangiabadi a,, Hamid Reza

More information

Frame Diagonalization of Matrices

Frame Diagonalization of Matrices Frame Diagonalization of Matrices Fumiko Futamura Mathematics and Computer Science Department Southwestern University 00 E University Ave Georgetown, Texas 78626 U.S.A. Phone: + (52) 863-98 Fax: + (52)

More information

The Hermitian R-symmetric Solutions of the Matrix Equation AXA = B

The Hermitian R-symmetric Solutions of the Matrix Equation AXA = B International Journal of Algebra, Vol. 6, 0, no. 9, 903-9 The Hermitian R-symmetric Solutions of the Matrix Equation AXA = B Qingfeng Xiao Department of Basic Dongguan olytechnic Dongguan 53808, China

More information

JOS M.F. TEN BERGE SIMPLICITY AND TYPICAL RANK RESULTS FOR THREE-WAY ARRAYS

JOS M.F. TEN BERGE SIMPLICITY AND TYPICAL RANK RESULTS FOR THREE-WAY ARRAYS PSYCHOMETRIKA VOL. 76, NO. 1, 3 12 JANUARY 2011 DOI: 10.1007/S11336-010-9193-1 SIMPLICITY AND TYPICAL RANK RESULTS FOR THREE-WAY ARRAYS JOS M.F. TEN BERGE UNIVERSITY OF GRONINGEN Matrices can be diagonalized

More information

Generic properties of Symmetric Tensors

Generic properties of Symmetric Tensors 2006 1/48 PComon Generic properties of Symmetric Tensors Pierre COMON - CNRS other contributors: Bernard MOURRAIN INRIA institute Lek-Heng LIM, Stanford University 2006 2/48 PComon Tensors & Arrays Definitions

More information

JOINT TOPOLOGICAL ZERO DIVISORS FOR A REAL BANACH ALGEBRA H. S. Mehta 1, R.D.Mehta 2, A. N. Roghelia 3

JOINT TOPOLOGICAL ZERO DIVISORS FOR A REAL BANACH ALGEBRA H. S. Mehta 1, R.D.Mehta 2, A. N. Roghelia 3 Mathematics Today Vol.30 June & December 2014; Published in June 2015) 54-58 ISSN 0976-3228 JOINT TOPOLOGICAL ZERO DIVISORS FOR A REAL BANACH ALGEBRA H. S. Mehta 1, R.D.Mehta 2, A. N. Roghelia 3 1,2 Department

More information

MAT 5330 Algebraic Geometry: Quiver Varieties

MAT 5330 Algebraic Geometry: Quiver Varieties MAT 5330 Algebraic Geometry: Quiver Varieties Joel Lemay 1 Abstract Lie algebras have become of central importance in modern mathematics and some of the most important types of Lie algebras are Kac-Moody

More information

The Solvability Conditions for the Inverse Eigenvalue Problem of Hermitian and Generalized Skew-Hamiltonian Matrices and Its Approximation

The Solvability Conditions for the Inverse Eigenvalue Problem of Hermitian and Generalized Skew-Hamiltonian Matrices and Its Approximation The Solvability Conditions for the Inverse Eigenvalue Problem of Hermitian and Generalized Skew-Hamiltonian Matrices and Its Approximation Zheng-jian Bai Abstract In this paper, we first consider the inverse

More information