Structured Low-Rank Approximation with Missing Data

@article{Markovsky2013StructuredLA,
  title={Structured Low-Rank Approximation with Missing Data},
  author={I. Markovsky and K. Usevich},
  journal={SIAM J. Matrix Anal. Appl.},
  year={2013},
  volume={34},
  pages={814-830}
}
We consider low-rank approximation of affinely structured matrices with missing elements. The method proposed is based on reformulation of the problem as inner and outer optimization. The inner minimization is a singular linear least-norm problem and admits an analytic solution. The outer problem is a nonlinear least-squares problem and is solved by local optimization methods: minimization subject to quadratic equality constraints and unconstrained minimization with regularized cost function… Expand
Regularized structured low-rank approximation with applications
TLDR
The proposed local optimization algorithm is able to solve the weighted structured low-rank approximation problem, as well as to deal with the cases of missing or fixed elements, and is designed to address the case of small targeted rank. Expand
Factorization Approach to Structured Low-Rank Approximation with Applications
TLDR
The proposed local optimization algorithm is able to solve the weighted structured low-rank approximation problem, as well as to deal with the cases of missing or fixed elements and compared to existing approaches on numerical examples of system identification, approximate greatest common divisor problem, and symmetric tensor decomposition. Expand
Using Hankel Structured Low-Rank Approximation for Sparse Signal Recovery
TLDR
This paper considers a modified Hankel structured low-rank approximation problem where the observed data is a linear transformation of a system’s trajectory with reduced dimension and proposes a solution methods based on the variable projections principle. Expand
Recent progress on variable projection methods for structured low-rank approximation
  • I. Markovsky
  • Mathematics, Computer Science
  • Signal Process.
  • 2014
TLDR
This paper gives an overview of recent progress in efficient local optimization algorithms for solving weighted mosaic-Hankel structured low-rank approximation problems. Expand
Robust Structured and Unstructured Low-Rank Approximation on the Grassmannian
TLDR
Algorithms for Robust PCA, Robust Subspace Tracking and Robust Structured Low-Rank Approximation are discussed, which involve optimization on the Grassmannian, the manifold of low-dimensional subspaces. Expand
On Low-Rank Hankel Matrix Denoising
TLDR
A novel approach is proposed to solve the low-rank Hankel matrix denoising problem by using an iterative algorithm in structured low- rank approximation modified with data-driven singular value shrinkage. Expand
Quasi-Hankel low-rank matrix completion: a convex relaxation
TLDR
This contribution analyzes the case of structured matrices with fixed pattern of missing values, in particular, the cases of Hankel and quasi-Hankel matrix completion, which appears as a subproblem in the computation of symmetric tensor canonical polyadic decomposition. Expand
Tensor Low Multilinear Rank Approximation by Structured Matrix Low-Rank Approximation *
We present a new connection between higherorder tensors and affinely structured matrices, in the context of low-rank approximation. In particular, we show that the tensor low multilinear rankExpand
Hankel Low-Rank Matrix Completion: Performance of the Nuclear Norm Relaxation
  • K. Usevich, P. Comon
  • Mathematics, Computer Science
  • IEEE Journal of Selected Topics in Signal Processing
  • 2016
TLDR
This contribution analyzes the case of structured matrices with a fixed pattern of missing values, namely, the Hankel matrix completion, and extends existing results on completion of rank-one real Hankel matrices to completion ofRank-r complex Hankel Matrices. Expand
Structured low-rank matrix learning: algorithms and applications
TLDR
This work considers the problem of learning a low-rank matrix, constrained to lie in a linear subspace, and introduces a novel factorization for modeling such matrices, and formulate the optimization problem on the Riemannian spectrahedron manifold. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 55 REFERENCES
Structured low-rank approximation and its applications
TLDR
This work outlines applications in system theory (approximate realization, model reduction, output error, and errors-in-variables identification), signal processing, signal processing (harmonic retrieval, sum-of-damped exponentials, and finite impulse response modeling), and computer algebra (Approximate common divisor). Expand
Solving a low-rank factorization model for matrix completion by a nonlinear successive over-relaxation algorithm
TLDR
A low-rank factorization model is proposed and a nonlinear successive over-relaxation (SOR) algorithm is constructed that only requires solving a linear least squares problem per iteration to improve the capacity of solving large-scale problems. Expand
Software for weighted structured low-rank approximation
TLDR
A software package is presented that computes locally optimal solutions to low-rank approximation problems with the following features:osaic Hankel structure constraint on the approximating matrix, weighted 2-norm approximation criterion, and linear constraints on an approximation matrix's left kernel basis. Expand
Variable projection for affinely structured low-rank approximation in weighted 2-norms
The structured low-rank approximation problem for general affine structures, weighted 2-norms and fixed elements is considered. The variable projection principle is used to reduce the dimensionalityExpand
Low-Rank Matrix Approximation with Weights or Missing Data Is NP-Hard
TLDR
This paper proves that computing an optimal WLRA is NP-hard, already when a rank-one approximation is sought, and shows that it is hard to compute approximate solutions to the WL RA problem with some prescribed accuracy. Expand
Overview of total least-squares methods
TLDR
It is explained how special structure of the weight matrix and the data matrix can be exploited for efficient cost function and first derivative computation that allows to obtain computationally efficient solution methods. Expand
Block-Toeplitz/Hankel Structured Total Least Squares
A structured total least squares problem is considered in which the extended data matrix is partitioned into blocks and each of the blocks is block-Toeplitz/Hankel structured, unstructured, or exact.Expand
Fast low-rank modifications of the thin singular value decomposition
This paper develops an identity for additive modivations of a singular value decomposition (SVD) to reflect updates, downdates, shifts, and edits of the data matrix. This sets the stage for fast andExpand
Algorithms and Literate Programs for Weighted Low-Rank Approximation with Missing Data
TLDR
Alternating projections and variable projections methods for solving the resulting problem are outlined and implemented in a literate programming style, using Matlab/Octave’s scripting language. Expand
Low Rank Approximation - Algorithms, Implementation, Applications
  • I. Markovsky
  • Computer Science
  • Communications and Control Engineering
  • 2012
TLDR
Data Approximation by Low-complexity Models details the theory, algorithms, and applications of structured low-rank approximation and describes the applications including: system and control theory; signal processing; computer algebra for approximate factorization and common divisor computation; computer vision for image deblurring and segmentation. Expand
...
1
2
3
4
5
...