Abstract | ||
---|---|---|
Robust matrix completion allows for estimating a low-rank matrix based on a subset of its entries, even in presence of impulsive noise and outliers. We explore recent progress in the theoretical analysis of non-convex low-rank factorization problems to develop a robust approach that is based on a fast factorization method. We propose an algorithm that uses joint regression and scale estimation to compute the estimates. We prove that our algorithm converges to a global minimum with random initialization. An example function for which the guarantees hold is the pseudo-Huber function. In simulations, the proposed approach is compared to state-of the art robust and nonrobust methods. In addition, its applicability to image inpainting and occlusion removal is demonstrated. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/MLSP49062.2020.9231573 | 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing (MLSP) |
Keywords | DocType | ISSN |
Matrix completion,optimality,low-rank factorization,robustness,image inpainting,occlusion removal | Conference | 1551-2541 |
ISBN | Citations | PageRank |
978-1-7281-6663-6 | 0 | 0.34 |
References | Authors | |
10 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Felicia Ruppel | 1 | 0 | 0.34 |
Michael Muma | 2 | 144 | 19.51 |
Abdelhak M. Zoubir | 3 | 1036 | 148.03 |