Task-Driven Uncertainty Quantification in Inverse Problems via Conformal Prediction
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910523633696768 |
|---|---|
| author | Wen, Jeffrey Ahmad, Rizwan Schniter, Philip |
| author_facet | Wen, Jeffrey Ahmad, Rizwan Schniter, Philip |
| contents | In imaging inverse problems, one seeks to recover an image from missing/corrupted measurements. Because such problems are ill-posed, there is great motivation to quantify the uncertainty induced by the measurement-and-recovery process. Motivated by applications where the recovered image is used for a downstream task, such as soft-output classification, we propose a task-centered approach to uncertainty quantification. In particular, we use conformal prediction to construct an interval that is guaranteed to contain the task output from the true image up to a user-specified probability, and we use the width of that interval to quantify the uncertainty contributed by measurement-and-recovery. For posterior-sampling-based image recovery, we construct locally adaptive prediction intervals. Furthermore, we propose to collect measurements over multiple rounds, stopping as soon as the task uncertainty falls below an acceptable level. We demonstrate our methodology on accelerated magnetic resonance imaging (MRI): https://github.com/jwen307/TaskUQ. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_18527 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Task-Driven Uncertainty Quantification in Inverse Problems via Conformal Prediction Wen, Jeffrey Ahmad, Rizwan Schniter, Philip Computer Vision and Pattern Recognition Image and Video Processing In imaging inverse problems, one seeks to recover an image from missing/corrupted measurements. Because such problems are ill-posed, there is great motivation to quantify the uncertainty induced by the measurement-and-recovery process. Motivated by applications where the recovered image is used for a downstream task, such as soft-output classification, we propose a task-centered approach to uncertainty quantification. In particular, we use conformal prediction to construct an interval that is guaranteed to contain the task output from the true image up to a user-specified probability, and we use the width of that interval to quantify the uncertainty contributed by measurement-and-recovery. For posterior-sampling-based image recovery, we construct locally adaptive prediction intervals. Furthermore, we propose to collect measurements over multiple rounds, stopping as soon as the task uncertainty falls below an acceptable level. We demonstrate our methodology on accelerated magnetic resonance imaging (MRI): https://github.com/jwen307/TaskUQ. |
| title | Task-Driven Uncertainty Quantification in Inverse Problems via Conformal Prediction |
| topic | Computer Vision and Pattern Recognition Image and Video Processing |
| url | https://arxiv.org/abs/2405.18527 |