Optimality of Approximate Message Passing Algorithms for Spiked Matrix Models with Rotationally Invariant Noise
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911139684679680 |
|---|---|
| author | Dudeja, Rishabh Liu, Songbin Ma, Junjie |
| author_facet | Dudeja, Rishabh Liu, Songbin Ma, Junjie |
| contents | We study the problem of estimating a rank one signal matrix from an observed matrix generated by corrupting the signal with additive rotationally invariant noise. We develop a new class of approximate message-passing algorithms for this problem and provide a simple and concise characterization of their dynamics in the high-dimensional limit. At each iteration, these algorithms exploit prior knowledge about the noise structure by applying a non-linear matrix denoiser to the eigenvalues of the observed matrix and prior information regarding the signal structure by applying a non-linear iterate denoiser to the previous iterates generated by the algorithm. We exploit our result on the dynamics of these algorithms to derive the optimal choices for the matrix and iterate denoisers. We show that the resulting algorithm achieves the smallest possible asymptotic estimation error among a broad class of iterative algorithms under a fixed iteration budget. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_18081 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Optimality of Approximate Message Passing Algorithms for Spiked Matrix Models with Rotationally Invariant Noise Dudeja, Rishabh Liu, Songbin Ma, Junjie Statistics Theory Information Theory Probability Machine Learning We study the problem of estimating a rank one signal matrix from an observed matrix generated by corrupting the signal with additive rotationally invariant noise. We develop a new class of approximate message-passing algorithms for this problem and provide a simple and concise characterization of their dynamics in the high-dimensional limit. At each iteration, these algorithms exploit prior knowledge about the noise structure by applying a non-linear matrix denoiser to the eigenvalues of the observed matrix and prior information regarding the signal structure by applying a non-linear iterate denoiser to the previous iterates generated by the algorithm. We exploit our result on the dynamics of these algorithms to derive the optimal choices for the matrix and iterate denoisers. We show that the resulting algorithm achieves the smallest possible asymptotic estimation error among a broad class of iterative algorithms under a fixed iteration budget. |
| title | Optimality of Approximate Message Passing Algorithms for Spiked Matrix Models with Rotationally Invariant Noise |
| topic | Statistics Theory Information Theory Probability Machine Learning |
| url | https://arxiv.org/abs/2405.18081 |