Name
Title | ||
|---|---|---|
Regularization-free estimation in trace regression with symmetric positive semidefinite matrices. |
Abstract | ||
|---|---|---|
Trace regression models have received considerable attention in the context of matrix completion, quantum state tomography, and compressed sensing. Estimation of the underlying matrix from regularization-based approaches promoting low-rankedness, notably nuclear norm regularization, have enjoyed great popularity. In this paper, we argue that such regularization may no longer be necessary if the underlying matrix is symmetric positive semidefinite (spd) and the design satisfies certain conditions. In this situation, simple least squares estimation subject to an spd constraint may perform as well as regularization-based approaches with a proper choice of regularization parameter, which entails knowledge of the noise level and/or tuning. By contrast, constrained least squares estimation comes without any tuning parameter and may hence be preferred due to its simplicity. |
Year | Venue | Field |
|---|---|---|
2015 | Annual Conference on Neural Information Processing Systems | Least squares,Mathematical optimization,Matrix completion,Matrix (mathematics),Positive-definite matrix,Matrix norm,Regularization (mathematics),Artificial intelligence,Machine learning,Mathematics,Compressed sensing,Regularization perspectives on support vector machines |
DocType | Volume | ISSN |
Journal | abs/1504.06305 | 1049-5258 |
Citations | PageRank | References |
1 | 0.36 | 10 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Martin Slawski | 1 | 16 | 6.14 |
| Ping Li | 2 | 1672 | 127.72 |
| Matthias Hein | 3 | 663 | 62.80 |