QUTCC:面向成像逆问题的分位数不确定性训练与保形校准方法
QUTCC: Quantile Uncertainty Training and Conformal Calibration for Imaging Inverse Problems
AI 导读
QUTCC 是一种面向成像逆问题的不确定性估计方法,采用带分位数嵌入的 U-Net 架构学习完整的条件分位数分布,并在保形校准时实现非线性、空间自适应的缩放。该方法在去噪、加速磁共振成像和定量相位显微等任务上,能在相同覆盖水平下产生比已有保形方法更紧的不确定性区间,还可预测逐像素条件概率密度,且高不确定性区域有时能帮助定位模型预测中的幻觉。
正文 · 原文
Abstract:While deep learning offers tremendous promise for scientific and medical imaging, any failures and hallucinations (predictions that do not coincide with reality) are hard to pinpoint and can have serious downstream consequences. Uncertainty estimation techniques, such as conformal prediction, can help by predicting statistically valid error bars for a model's prediction. However, popular conformal prediction methods were not designed for high-dimensional image-valued problems and do not take into account spatial correlations within an image during conformal calibration, resulting in larger-than-necessary uncertainty intervals. We propose a practical simultaneous quantile regression method that enables non-linear, spatially-adaptive scaling during conformal calibration. Our method, QUTCC uses a U-Net architecture with a quantile embedding to learn a full conditional quantile distribution during training, and then leverages this non-linear, learned function for spatially-adaptive conformal calibration. At test time, our method can efficiently estimate uncertainty intervals with pixel-marginal coverage guarantees. In addition, QUTCC can also predict pixel-wise conditional probability density estimates without any built-in distributional assumptions. We evaluate our method on several denoising problems, accelerated magnetic resonance imaging, and quantitative phase microscopy. Our method consistently produces tighter uncertainty intervals than prior conformal methods at the same coverage level, can predict plausible conditional distributions for different tasks, and in some cases, high-uncertainty regions can help us locate hallucinations in a model's prediction.
| Comments: | for associated project page and code, see this https URL |
| Subjects: | Image and Video Processing (eess.IV); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) |
| Cite as: | arXiv:2507.14760 [eess.IV] |
| (or arXiv:2507.14760v3 [eess.IV] for this version) | |
| https://doi.org/10.48550/arXiv.2507.14760 arXiv-issued DOI via DataCite |
Submission history
From: Cassandra Ye [view email]
[v1]
Sat, 19 Jul 2025 21:44:14 UTC (1,482 KB)
[v2]
Sun, 24 May 2026 19:36:49 UTC (2,815 KB)
[v3]
Tue, 17 Nov 2026 03:02:25 UTC (3,941 KB)
来源:arXiv cs.AI · arxiv.org