Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Calibrated Inference: Statistical Inference that Accounts for Both Sampling Uncertainty and Distributional Uncertainty
Yujin Jeong, Dominik Rothenhäusler
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 |
|---|---|---|---|
| add calibration (estimator-to-estimator scaling of confidence intervals) to the aggregated regression estimators | linear regression covariate ranking · UCI student performance, K=10 | top-1 ranking set similarity: 0.102 -> 0.210 | Improved |
| add calibration (estimator-to-estimator scaling of confidence intervals) to the aggregated regression estimators | linear regression covariate ranking · UCI student performance, K=20 | top-1 ranking set similarity: 0.090 -> 0.235 | Improved |
| add calibration to the aggregated estimators | linear regression covariate ranking · UCI student performance, K=10 | top-4 ranking set similarity: 0.648 -> 0.658 | Improved |
| increase number of covariate sets from K=10 to K=20 for calibrated method (ablation) | calibrated inference · UCI student performance | top-1 ranking set similarity: 0.210 -> 0.235 | Improved |
| add highly correlated estimators (K=8 adjustment sets instead of K=6) (ablation) | calibrated confidence intervals · simulated perturbed SCM | marginal coverage | Hurt |
| use calibrated intervals instead of non-calibrated intervals | individual regression-adjusted estimators · simulated perturbed SCM (n,m=200,500,1000) | marginal coverage | Improved |
| use calibrated intervals instead of non-calibrated intervals | individual regression-adjusted estimators · simulated perturbed SCM | confidence interval length | Improved |
Closest papers by meaning
- Not all distributional shifts are equal: Fine-grained robust conformal inference (ICML 2024 · similarity 0.85)
- Calibration without labels in multiple testing (NeurIPS 2026 · similarity 0.84)
- Trust, but Don’t Verify: Epistemic Blind Spots in LLM Source Evaluation (NeurIPS 2026 · similarity 0.84)
- Asymptotically Log-Optimal Bayes-Assisted Confidence Sequences for Bounded Means (NeurIPS 2026 · similarity 0.82)
- Improving Perturbation-based Explanations by Understanding the Role of Uncertainty Calibration (NeurIPS 2025 · similarity 0.82)
- Experts Don't Cheat: Learning What You Don't Know By Predicting Pairs (ICML 2024 · similarity 0.82)
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.
