Successive Rank-One Approximations for Nearly Orthogonally Decomposable Symmetric Tensors
Authors:
- Cun MuCMCun Mu
- Daniel HsuDHDaniel Hsu
- Donald GoldfarbDGDonald Goldfarb
- 1 January 2015
- journal article
- Published by Society for Industrial & Applied Mathematics (SIAM) in SIAM Journal on Matrix Analysis and Applications
- Vol. 36 (4) , 1638-1659
Abstract
Many idealized problems in signal processing, machine learning, and statistics can be reduced to the problem of finding the symmetric canonical decomposition of an underlying symmetric and orthogonally decomposable (SOD) tensor. Drawing inspiration from the matrix case, the successive rank-one approximation (SROA) scheme has been proposed and shown to yield this tensor decomposition exactly, and a plethora of numerical methods have thus been developed for the tensor rank-one approximation problem. In practice, however, the inevitable errors (say) from estimation, computation, and modeling necessitate that the input tensor can only be assumed to be a nearly SOD tensor---i.e., a symmetric tensor slightly perturbed from the underlying SOD tensor. This paper shows that even in the presence of perturbation, SROA can still robustly recover the symmetric canonical decomposition of the underlying tensor. It is shown that when the perturbation error is small enough, the approximation errors do not accumulate with ...Keywords
All Related Versions
This publication has 35 references indexed in Scilit:
- A sequential subspace projection method for extreme Z‐eigenvalues of supersymmetric tensorsNumerical Linear Algebra with Applications, 2014
- Most Tensor Problems Are NP-HardJournal of the ACM, 2013
- An unconstrained optimization approach for finding real eigenvalues of even order symmetric tensorsNumerical Algebra, Control & Optimization, 2013
- Maximum Block Improvement and Polynomial OptimizationSIAM Journal on Optimization, 2012
- GloptiPoly 3: moments, optimization and semidefinite programmingOptimization Methods and Software, 2009
- Tensor Rank and the Ill-Posedness of the Best Low-Rank Approximation ProblemSIAM Journal on Matrix Analysis and Applications, 2008
- On the Best Rank-1 and Rank-(R1 ,R2 ,. . .,RN) Approximation of Higher-Order TensorsSIAM Journal on Matrix Analysis and Applications, 2000
- Independent component analysis, A new concept?Signal Processing, 1994
- The Rotation of Eigenvectors by a Perturbation. IIISIAM Journal on Numerical Analysis, 1970
- The Approximation of One Matrix by Another of Lower RankPsychometrika, 1936