Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Image-to-Image Regression with Distribution-Free Uncertainty Quantification and Applications in Imaging
Anastasios N Angelopoulos, Amit Pal Kohli, Stephen Bates, Michael Jordan, Jitendra Malik, Thayer Alshaabi, Srigokul Upadhyayula, Yaniv Romano
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use pixelwise quantile regression instead of the other uncertainty heuristics | U-Net image-to-image regression · Quantitative phase microscopy (BSCCM leukocytes) | average interval size | Improved |
| use pixelwise quantile regression instead of the other uncertainty heuristics | U-Net image-to-image regression · Quantitative phase microscopy (BSCCM leukocytes) | size-stratified risk | Improved |
| use pixelwise quantile regression instead of the other uncertainty heuristics | U-Net image-to-image regression · Quantitative phase microscopy (BSCCM leukocytes) | point-prediction MSE | No clear effect |
| use the Gaussian uncertainty heuristic | U-Net image-to-image regression · Quantitative phase microscopy (BSCCM leukocytes) | training convergence | Hurt |
| use the softmax uncertainty heuristic | U-Net image-to-image regression · Quantitative phase microscopy (BSCCM leukocytes) | interval adaptivity | Hurt |
| use pixelwise quantile regression instead of the other uncertainty heuristics | U-Net image-to-image regression · FastMRI knee reconstruction | interval size | Improved |
| use pixelwise quantile regression instead of the other uncertainty heuristics | U-Net image-to-image regression · FastMRI knee reconstruction | size-stratified risk | Improved |
| use the softmax uncertainty heuristic | U-Net image-to-image regression · FastMRI knee reconstruction | interval-size adaptivity | Hurt |
| use pixelwise quantile regression | U-Net image-to-image regression · FastMRI knee reconstruction | empirical risk | Improved |
| use pixelwise quantile regression | U-Net image-to-image regression · TEMCA2 Drosophila brain super-resolution | size-stratified risk | Mixed |
| use Gaussian negative log-likelihood training | joint point prediction and uncertainty heuristic · Image-to-image regression experiments | prediction quality | Hurt |
Closest papers by meaning
- QUTCC: Quantile Uncertainty Training and Conformal Calibration for Imaging Inverse Problems (NeurIPS 2026 · similarity 0.85)
- Non-Asymptotic Uncertainty Quantification in High-Dimensional Learning (NeurIPS 2024 · similarity 0.84)
- Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstruction (NeurIPS 2024 · similarity 0.84)
- The Perception-Robustness Tradeoff in Deterministic Image Restoration (ICML 2024 · similarity 0.84)
- Entropy Minimization without Model Collapse: Mitigating Prediction Bias in Medical Imaging (NeurIPS 2026 · similarity 0.83)
- Consistency Posterior Sampling for Diverse Image Synthesis (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
