Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Training-Free Uncertainty Estimation for Dense Regression: Sensitivity as a Surrogate
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 infer-transformation instead of MC-dropout² for uncertainty estimation | SRGAN · Set14 | mean AUSE: 0.007 -> 0.006 | Helped |
| Use infer-transformation instead of MC-dropout¹ for uncertainty estimation | SRGAN · Set14 | mean correlation: 0.943 -> 0.931 | Hurt |
| Use infer-transformation instead of LLM for uncertainty estimation | SRGAN · Set14 | NLL: 1.320 -> 9.332 | Hurt |
| Use infer-noise instead of MC-dropout¹ for uncertainty estimation | FCRN · NYU Depth Dataset V2 | mean correlation: 0.473 -> 0.708 | Helped |
| Use infer-noise instead of MC-dropout¹ for uncertainty estimation | FCRN · NYU Depth Dataset V2 | NLL: 12.365 -> 1.006 | Helped |
| Use entropy instead of MC-dropout to estimate uncertainty | Densenet · CamVid | mean correlation: 0.718 -> 0.928 | Helped |
| Use entropy instead of MC-dropout to estimate uncertainty | UNET · SNEMI3D | mean correlation: 0.881 -> 0.964 | Helped |
| Use entropy instead of MC-dropout to estimate uncertainty | Resnet · CIFAR100 | mean correlation: 0.537 -> 0.669 | Helped |
| Select training examples using high uncertainty rather than randomly | Active learning for SR · Active learning (SR) | SSIM: 0.773 -> 0.790 | Helped |
| Select training examples using high uncertainty rather than randomly | Active learning for SR · Active learning (SR) | PSNR: 23.616 -> 23.942 | Helped |
| Select training examples using high uncertainty rather than randomly | Active learning for SR · Active learning (SR) | L1: 12.344 -> 11.497 | Helped |
| Keep perturbations tolerable, with less than 20% relative performance drop (ablation) | Training-free uncertainty methods · Super resolution and depth estimation | correlation between uncertainty and error | Helped |
Closest papers by meaning
- Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles (NeurIPS 2017 · similarity 0.90)
- Uncertainty Quantification for Deep Regression using Contextualised Normalizing Flows (NeurIPS 2025 · similarity 0.89)
- SURE: SUrvey REcipes for building reliable and robust deep networks (CVPR 2024 · similarity 0.88)
- Density-Softmax: Efficient Test-time Model for Uncertainty Estimation and Robustness under Distribution Shifts (ICML 2024 · similarity 0.88)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.87)
- Make Me a BNN: A Simple Strategy for Estimating Bayesian Uncertainty from Pre-trained Models (CVPR 2024 · similarity 0.87)
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.