Subtracting a best rank-1 approximation may increase tensor rank
From MaRDI portal
Publication:5962281
DOI10.1016/J.LAA.2010.06.027zbMATH Open1198.15018arXiv0906.0483OpenAlexW2569661359MaRDI QIDQ5962281FDOQ5962281
Publication date: 21 September 2010
Published in: Linear Algebra and its Applications (Search for Journal in Brave)
Abstract: It has been shown that a best rank-R approximation of an order-k tensor may not exist when R>1 and k>2. This poses a serious problem to data analysts using tensor decompositions. It has been observed numerically that, generally, this issue cannot be solved by consecutively computing and subtracting best rank-1 approximations. The reason for this is that subtracting a best rank-1 approximation generally does not decrease tensor rank. In this paper, we provide a mathematical treatment of this property for real-valued 2x2x2 tensors, with symmetric tensors as a special case. Regardless of the symmetry, we show that for generic 2x2x2 tensors (which have rank 2 or 3), subtracting a best rank-1 approximation results in a tensor that has rank 3 and lies on the boundary between the rank-2 and rank-3 sets. Hence, for a typical tensor of rank 2, subtracting a best rank-1 approximation increases the tensor rank.
Full work available at URL: https://arxiv.org/abs/0906.0483
Recommendations
- Subtracting a best rank‐1 approximation from p × p × 2(p≥2) tensors
- On best rank-\(2\) and rank-\((2,2,2)\) approximations of order-\(3\) tensors
- On the Best Rank-1 and Rank-(R1 ,R2 ,. . .,RN) Approximation of Higher-Order Tensors
- Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation Problem
- The Best rank-\((R_1,R_2,R_3)\) approximation of tensors by means of a geometric Newton method
data analysislow-rank approximationsymmetric tensorstensor ranktensor decompositionmulti-wayparafaccandecomp
Multilinear algebra, tensor calculus (15A69) Vector spaces, linear dependence, rank, lineability (15A03)
Cites Work
- Independent component analysis, a new concept?
- Analysis of individual differences in multidimensional scaling via an \(n\)-way generalization of ``Eckart-Young decomposition
- Applied Multiway Data Analysis
- Tensor Decompositions and Applications
- Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics
- Explicit Candecomp/Parafac solutions for a contrived 2\(\times 2\times 2\) array of rank three
- On the Best Rank-1 and Rank-(R1 ,R2 ,. . .,RN) Approximation of Higher-Order Tensors
- Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation Problem
- On the best rank-1 approximation of higher-order supersymmetric tensors
- Title not available (Why is that?)
- Rank and eigenvalues of a supersymmetric tensor, the multivariate homogeneous polynomial and the algebraic hypersurface it defines
- Decomposition of quantics in sums of powers of linear forms
- Symmetric Tensors and Symmetric Tensor Rank
- On the best rank-1 approximation to higher-order symmetric tensors
- Rank-one approximation to high order tensors
- Degeneracy in Candecomp/Parafac explained for \(p\times p\times 2\) arrays of rank \(p+1\) or higher
- The invariant theory of binary forms
- On the non-existence of optimal solutions and the occurrence of ``degeneracy in the CANDECOMP/PARAFAC model
- Simplicity of core arrays in three-way principal component analysis and the typical rank of \(p\times q\times 2\) arrays
- Optimal Evaluation of Pairs of Bilinear Forms
- Rank and optimal computation of generic tensors
- Typical rank and indscal dimensionality for symmetric three-way arrays of order \(I\times 2\times 2\) or \(I\times 3\times 3\)
- Tensor-based techniques for the blind separation of DS-CDMA signals
- Degeneracy in Candecomp/Parafac and Indscal explained for several three-sliced arrays with a two-valued typical rank
- Low-Rank Approximation of Generic $p \timesq \times2$ Arrays and Diverging Components in the Candecomp/Parafac Model
- On Kruskal's uniqueness condition for the Candecomp/Parafac decomposition
- PARAFAC-based unified tensor modeling for wireless communication systems with application to blind multiuser equalization
- A Method to Avoid Diverging Components in the Candecomp/Parafac Model for Generic $I\timesJ\times2$ Arrays
Cited In (14)
- The Euclidean distance degree of an algebraic variety
- Cayley's hyperdeterminant: A combinatorial approach via representation theory
- Successive Rank-One Approximations for Nearly Orthogonally Decomposable Symmetric Tensors
- On the hyperdeterminant for 2×2×3 arrays
- Rank of a tensor and quantum entanglement
- A very brief introduction to nonnegative tensors from the geometric viewpoint
- A Higher Order Unscented Transform
- Tensor Decompositions for Learning Latent Variable Models (A Survey for ALT)
- Subtracting a best rank‐1 approximation from p × p × 2(p≥2) tensors
- A literature survey of low-rank tensor approximation techniques
- Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation Problem
- Factorization strategies for third-order tensors
- A Constructive Algorithm for Decomposing a Tensor into a Finite Sum of Orthonormal Rank-1 Terms
- On the Uniqueness and Perturbation to the Best Rank-One Approximation of a Tensor
This page was built for publication: Subtracting a best rank-1 approximation may increase tensor rank
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5962281)
