UNSURE: self-supervised learning with Unknown Noise level and Stein's Unbiased Risk Estimate
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909487114223616 |
|---|---|
| author | Tachella, Julián Davies, Mike Jacques, Laurent |
| author_facet | Tachella, Julián Davies, Mike Jacques, Laurent |
| contents | Recently, many self-supervised learning methods for image reconstruction have been proposed that can learn from noisy data alone, bypassing the need for ground-truth references. Most existing methods cluster around two classes: i) Stein's Unbiased Risk Estimate (SURE) and similar approaches that assume full knowledge of the noise distribution, and ii) Noise2Self and similar cross-validation methods that require very mild knowledge about the noise distribution. The first class of methods tends to be impractical, as the noise level is often unknown in real-world applications, and the second class is often suboptimal compared to supervised learning. In this paper, we provide a theoretical framework that characterizes this expressivity-robustness trade-off and propose a new approach based on SURE, but unlike the standard SURE, does not require knowledge about the noise level. Throughout a series of experiments, we show that the proposed estimator outperforms other existing self-supervised methods on various imaging inverse problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_01985 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | UNSURE: self-supervised learning with Unknown Noise level and Stein's Unbiased Risk Estimate Tachella, Julián Davies, Mike Jacques, Laurent Machine Learning Signal Processing 68U10 I.4.5; I.2.10; G.3 Recently, many self-supervised learning methods for image reconstruction have been proposed that can learn from noisy data alone, bypassing the need for ground-truth references. Most existing methods cluster around two classes: i) Stein's Unbiased Risk Estimate (SURE) and similar approaches that assume full knowledge of the noise distribution, and ii) Noise2Self and similar cross-validation methods that require very mild knowledge about the noise distribution. The first class of methods tends to be impractical, as the noise level is often unknown in real-world applications, and the second class is often suboptimal compared to supervised learning. In this paper, we provide a theoretical framework that characterizes this expressivity-robustness trade-off and propose a new approach based on SURE, but unlike the standard SURE, does not require knowledge about the noise level. Throughout a series of experiments, we show that the proposed estimator outperforms other existing self-supervised methods on various imaging inverse problems. |
| title | UNSURE: self-supervised learning with Unknown Noise level and Stein's Unbiased Risk Estimate |
| topic | Machine Learning Signal Processing 68U10 I.4.5; I.2.10; G.3 |
| url | https://arxiv.org/abs/2409.01985 |