Noisy Matrix Completion Under Sparse Factor Models
From MaRDI portal
Abstract: This paper examines a general class of noisy matrix completion tasks where the goal is to estimate a matrix from observations obtained at a subset of its entries, each of which is subject to random noise or corruption. Our specific focus is on settings where the matrix to be estimated is well-approximated by a product of two (a priori unknown) matrices, one of which is sparse. Such structural models - referred to here as "sparse factor models" - have been widely used, for example, in subspace clustering applications, as well as in contemporary sparse modeling and dictionary learning tasks. Our main theoretical contributions are estimation error bounds for sparsity-regularized maximum likelihood estimators for problems of this form, which are applicable to a number of different observation noise or corruption models. Several specific implications are examined, including scenarios where observations are corrupted by additive Gaussian noise or additive heavier-tailed (Laplace) noise, Poisson-distributed observations, and highly-quantized (e.g., one-bit) observations. We also propose a simple algorithmic approach based on the alternating direction method of multipliers for these tasks, and provide experimental evidence to support our error analyses.
Cited in
(5)- Structured matrix estimation and completion
- Matrix completion from noisy entries
- A nonlinear matrix decomposition for mining the zeros of sparse data
- Low-rank matrix completion with Poisson observations via nuclear norm and total variation constraints
- Poisson tensor completion via nonconvex regularization and nonlocal self-similarity for multi-dimensional image recovery
This page was built for publication: Noisy Matrix Completion Under Sparse Factor Models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2976813)
