On Uniqueness and Computation of the Decomposition of a Tensor into Multilinear Rank-(1,L_r,L_r) Terms

DOI10.1137/18M1206849zbMATH Open1442.15042arXiv1808.02423OpenAlexW3022422766MaRDI QIDQ5113366FDOQ5113366

Ignat Domanov, Lieven De Lathauwer

Publication date: 4 June 2020

Published in: SIAM Journal on Matrix Analysis and Applications (Search for Journal in Brave)

Abstract: Canonical Polyadic Decomposition (CPD) represents a third-order tensor as the minimal sum of rank-1 terms. Because of its uniqueness properties the CPD has found many concrete applications in telecommunication, array processing, machine learning, etc. On the other hand, in several applications the rank-1 constraint on the terms is too restrictive. A multilinear rank-$(M,N,L)$ constraint (where a rank-1 term is the special case for which $M=N=L=1$) could be more realistic, while it still yields a decomposition with attractive uniqueness properties. In this paper we focus on the decomposition of a tensor $mathcalT$ into a sum of multilinear rank-$(1,L_r,L_r)$ terms, $r=1,\mathrm{...},R$. This particular decomposition type has already found applications in wireless communication, chemometrics and the blind signal separation of signals that can be modelled as exponential polynomials and rational functions. We find conditions on the terms which guarantee that the decomposition is unique and can be computed by means of the eigenvalue decomposition of a matrix even in the cases where none of the factor matrices has full column rank. We consider both the case where the decomposition is exact and the case where the decomposition holds only approximately. We show that in both cases the number of the terms $R$ and their "sizes" $L_1,\mathrm{...},L_R$ do not have to be known a priori and can be estimated as well. The conditions for uniqueness are easy to verify, especially for terms that can be considered "generic". In particular, we obtain the following two generalizations of a well known result on generic uniqueness of the CPD (i.e., the case $L_1=\mathrm{...}=L_R=1$): we show that the multilinear rank-$(1,L_r,L_r)$ decomposition of an $IimesJimesK$ tensor is generically unique if i) $L_1=\mathrm{...}=L_R=:L$ and $Rleqmin((J-L)(K-L),I)$ or if ii) $sum{L}_{R}leqmin((I-1)(J-1),K)$ and $Jgeqmax(L_i+L_j)$.

Full work available at URL: https://arxiv.org/abs/1808.02423

zbMATH Keywords

factor analysiseigenvalue decompositionmultilinear algebrauniquenesssignal separationmultilinear rankblock term decompositionthird-order tensor

Mathematics Subject Classification ID

Eigenvalues, singular values, and eigenvectors (15A18) Factorization of matrices (15A23) Multilinear algebra, tensor calculus (15A69)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Tensor Decompositions and Applications
• On the uniqueness of the canonical polyadic decomposition of third-order tensors. I: Basic results and uniqueness of one factor matrix
• On the uniqueness of the canonical polyadic decomposition of third-order tensors. II: Uniqueness of the overall decomposition
• Canonical polyadic decomposition of third-order tensors: reduction to generalized eigenvalue decomposition
• Generic Uniqueness Conditions for the Canonical Polyadic Decomposition and INDSCAL
• An Algorithm For Generic and Low-Rank Specific Identifiability of Complex Tensors
• A Link between the Canonical Decomposition in Multilinear Algebra and Simultaneous Matrix Diagonalization
• Algebraic geometry. An introduction. Transl. from the French by Catriona Maclean
• Decompositions of a Higher-Order Tensor in Block Terms—Part II: Definitions and Uniqueness
• Rank and optimal computation of generic tensors
• Refined methods for the identifiability of tensors
• An Algebraic Approach to Nonorthogonal General Joint Block Diagonalization
• Finding a low-rank basis in a matrix subspace
• Tensor Decomposition for Signal Processing and Machine Learning
• Effective Criteria for Specific Identifiability of Tensors and Forms
• Preconditioned optimization algorithms solving the problem of the non unitary joint block diagonalization: application to blind separation of convolutive mixtures
• Real identifiability vs. complex identifiability
• Pencil-Based Algorithms for Tensor Rank Decomposition are not Stable
• Blind Deconvolution of DS-CDMA Signals by Means of Decomposition in Rank-(1, L, L) Terms
• Coupled Canonical Polyadic Decompositions and (Coupled) Decompositions in Multilinear Rank- $(L_{r,n},L_{r,n},1)$ Terms---Part II: Algorithms
• On partial and generic uniqueness of block term tensor decompositions
• Blind Separation of Exponential Polynomials and the Decomposition of a Tensor in Rank-$(L_r,L_r,1)$ Terms
• Löwner-Based Blind Signal Separation of Rational Functions With Applications

Cited In (9)

• Decomposition of a tensor into multilinear rank-\((M_r,N_r,\cdot)\) terms
• Lower bounds on the rank and symmetric rank of real tensors
• From Computation to Comparison of Tensor Decompositions
• $(L_r,L_r,1)$-Decompositions, Sparse Component Analysis, and the Blind Separation of Sums of Exponentials
• Structured Rank-(r1, . . . , rd) Decomposition of Function-related Tensors in R_D
• Nonlinear algebra and applications
• Improved Uniqueness Conditions for Canonical Tensor Decompositions with Linearly Dependent Loadings
• Blind Separation of Exponential Polynomials and the Decomposition of a Tensor in Rank-$(L_r,L_r,1)$ Terms
• SOTT: Greedy Approximation of a Tensor as a Sum of Tensor Trains

Uses Software

• Tensorlab

   Recommendations

• On Uniqueness of the nth Order Tensor Decomposition into Rank-1 Terms with Linear Independence in One Mode 👍 👎
• Tensor decompositions in rank +1 👍 👎
• A Constructive Algorithm for Decomposing a Tensor into a Finite Sum of Orthonormal Rank-1 Terms 👍 👎
• On partial and generic uniqueness of block term tensor decompositions 👍 👎
• Rank-\(r\) decomposition of symmetric tensors 👍 👎
• Tensor decompositions and rank increment conjecture 👍 👎
• On Uniqueness of the Canonical Tensor Decomposition with Some Form of Symmetry 👍 👎
• On the Uniqueness and Perturbation to the Best Rank-One Approximation of a Tensor 👍 👎
• Structured Rank-(r1, . . . , rd) Decomposition of Function-related Tensors in R_D 👍 👎
• Rank properties and computational methods for orthogonal tensor decompositions 👍 👎

This page was built for publication: On Uniqueness and Computation of the Decomposition of a Tensor into Multilinear Rank-$(1,L_r,L_r)$ Terms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5113366)