Research map / Time series / Causal and generative time series
Building Calibrated Deep Models via Uncertainty Matching with Auxiliary Interval Predictors
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 |
|---|---|---|---|
| Apply uncertainty matching with Sigma Fit or IQR Fit | FCN regression model · Red Wine, White Wine, Parkinsons, Boston Housing | RMSE: Red Wine: 0.737 -> 0.581; White Wine: 0.782 -> 0.691; Parkinsons: 2.493 -> 2.017; Boston Housing: 4.025 -> 4.011 | Helped |
| Apply uncertainty matching | FCN regression model · Auto MPG | RMSE: 3.07 -> 3.18 | Hurt |
| Apply uncertainty matching with Sigma Fit or IQR Fit | LSTM forecasting model · NSW2013, Bike Sharing, TAS2016, Traffic, Beijing, Air Quality | RMSE: NSW2013: 124.59 -> 90.31; Bike Sharing: 64.38 -> 53.65; TAS2016: 22.81 -> 17.64; Traffic: 1.36E+08 -> 1.21E+08; Beijing: | Helped |
| Use IQR Fit instead of HNN | LSTM forecasting model · All six forecasting datasets | CE_0.95: NSW2013: 0.008 -> 0.003; Bike Sharing: 0.004 -> 0.002; TAS2016: 0.013 -> 0.005; Traffic: 0.014 -> 0.003; Beijing: 0.028 | Helped |
| Use IQR Fit instead of HNN | ResNet-18 object localization model · CUB | IoU: 0.468 -> 0.557 | Helped |
| Use IQR Fit instead of HNN | ResNet-18 object localization model · CUB bounding-box coordinates | CE_0.95: x: 0.11 -> 0.01; y: 0.14 -> 0.02; w: 0.12 -> 0.04; h: 0.15 -> 0.04 | Helped |
Closest papers by meaning
- High-Quality Prediction Intervals for Deep Learning: A Distribution-Free, Ensembled Approach (ICML 2018 · similarity 0.90)
- Uncertainty Quantification for Deep Regression using Contextualised Normalizing Flows (NeurIPS 2025 · similarity 0.86)
- Reassessing How to Compare and Improve the Calibration of Machine Learning Models (ICLR 2025 · similarity 0.86)
- Uncertainty Weighted Gradients for Model Calibration (CVPR 2025 · similarity 0.85)
- A Benchmark Study on Calibration (ICLR 2024 · similarity 0.85)
- Self-Calibrating Conformal Prediction (NeurIPS 2024 · similarity 0.85)
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.