Greedy approaches to symmetric orthogonal tensor decomposition

DOI10.1137/16M1087734zbMATH Open1386.15051arXiv1706.01169OpenAlexW2624256161MaRDI QIDQ4588941FDOQ4588941

Authors: Cun Mu, Daniel Hsu, Donald Goldfarb 

Publication date: 6 November 2017

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

Abstract: Finding the symmetric and orthogonal decomposition (SOD) of a tensor is a recurring problem in signal processing, machine learning and statistics. In this paper, we review, establish and compare the perturbation bounds for two natural types of incremental rank-one approximation approaches. Numerical experiments and open questions are also presented and discussed.

Full work available at URL: https://arxiv.org/abs/1706.01169

Recommendations

• Successive Rank-One Approximations for Nearly Orthogonally Decomposable Symmetric Tensors
• Symmetric rank-1 approximation of symmetric high-order tensors
• On the perturbation of rank-one symmetric tensors.
• Symmetric tensor decomposition
• Monotonically convergent algorithms for symmetric tensor approximation

zbMATH Keywords

perturbation analysistensor decompositionorthogonally decomposable tensorrank-one tensor approximation

Mathematics Subject Classification ID

Eigenvalues, singular values, and eigenvectors (15A18) Multilinear algebra, tensor calculus (15A69) Numerical linear algebra (65F99) Decomposition methods (49M27)

Cites Work

• The elements of statistical learning. Data mining, inference, and prediction
• Numerical Optimization
• Independent component analysis, a new concept?
• The Rotation of Eigenvectors by a Perturbation. III
• Orthogonal tensor decompositions
• Title not available (Why is that?)
• Maximum block improvement and polynomial optimization
• On the Best Rank-1 and Rank-(R₁ ,R₂ ,. . .,R_N) Approximation of Higher-Order Tensors
• Global optimization with polynomials and the problem of moments
• Semidefinite programming relaxations for semialgebraic problems
• On the best rank-1 approximation of higher-order supersymmetric tensors
• Shifted power method for computing tensor eigenpairs
• GloptiPoly 3: moments, optimization and semidefinite programming
• Tensor decompositions for learning latent variable models
• The best rank-1 approximation of a symmetric tensor and related spherical optimization problems
• Properties and methods for finding the best rank-one approximation to higher-order tensors
• Rank-one approximation to high order tensors
• Semidefinite Relaxations for Best Rank-1 Tensor Approximations
• On the successive supersymmetric rank-1 decomposition of higher-order supersymmetric tensors
• A note on semidefinite programming relaxations for polynomial optimization over a single sphere
• Title not available (Why is that?)
• An unconstrained optimization approach for finding real eigenvalues of even order symmetric tensors
• An approach to obtaining global extremums in polynomial mathematical programming problems
• A sequential subspace projection method for extreme Z-eigenvalues of supersymmetric tensors
• Successive Rank-One Approximations for Nearly Orthogonally Decomposable Symmetric Tensors

Cited In (3)

• Successive partial-symmetric rank-one algorithms for almost unitarily decomposable conjugate partial-symmetric tensors
• Title not available (Why is that?)
• Using negative curvature in solving nonlinear programs

Uses Software

• Sostools
• GloptiPoly

This page was built for publication: Greedy approaches to symmetric orthogonal tensor decomposition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4588941)