Guarantees for existence of a best canonical polyadic approximation of a noisy low-rank tensor

12/15/2021
βˆ™
by   Eric Evert, et al.
βˆ™
0
βˆ™

The canonical polyadic decomposition (CPD) of a low rank tensor plays a major role in data analysis and signal processing by allowing for unique recovery of underlying factors. However, it is well known that the low rank CPD approximation problem is ill-posed. That is, a tensor may fail to have a best rank R CPD approximation when R>1. This article gives deterministic bounds for the existence of best low rank tensor approximations over 𝕂=ℝ or 𝕂=β„‚. More precisely, given a tensor π’―βˆˆπ•‚^I Γ— I Γ— I of rank R ≀ I, we compute the radius of a Frobenius norm ball centered at 𝒯 in which best 𝕂-rank R approximations are guaranteed to exist. In addition we show that every 𝕂-rank R tensor inside of this ball has a unique canonical polyadic decomposition. This neighborhood may be interpreted as a neighborhood of "mathematical truth" in with CPD approximation and computation is well-posed. In pursuit of these bounds, we describe low rank tensor decomposition as a β€œjoint generalized eigenvalue" problem. Using this framework, we show that, under mild assumptions, a low rank tensor which has rank strictly greater than border rank is defective in the sense of algebraic and geometric multiplicities for joint generalized eigenvalues. Bounds for existence of best low rank approximations are then obtained by establishing perturbation theoretic results for the joint generalized eigenvalue problem. In this way we establish a connection between existence of best low rank approximations and the tensor spectral norm. In addition we solve a "tensor Procrustes problem" which examines orthogonal compressions for pairs of tensors. The main results of the article are illustrated by a variety of numerical experiments.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
βˆ™ 08/16/2022

Low Rank Tensor Decompositions and Approximations

There exist linear relations among tensor entries of low rank tensors. T...
research
βˆ™ 04/13/2020

Model-Free State Estimation Using Low-Rank Canonical Polyadic Decomposition

As electric grids experience high penetration levels of renewable genera...
research
βˆ™ 02/23/2022

Canonical Polyadic Decomposition via the generalized Schur decomposition

The canonical polyadic decomposition (CPD) is a fundamental tensor decom...
research
βˆ™ 12/15/2021

A recursive eigenspace computation for the Canonical Polyadic decomposition

The canonical polyadic decomposition (CPD) is a compact decomposition wh...
research
βˆ™ 05/29/2017

Successive Rank-One Approximations for Nearly Orthogonally Decomposable Symmetric Tensors

Many idealized problems in signal processing, machine learning and stati...
research
βˆ™ 01/22/2020

Rank Bounds for Approximating Gaussian Densities in the Tensor-Train Format

Low rank tensor approximations have been employed successfully, for exam...
research
βˆ™ 11/03/2021

Tensor Decomposition Bounds for TBM-Based Massive Access

Tensor-based modulation (TBM) has been proposed in the context of unsour...

Please sign up or login with your details

Forgot password? Click here to reset