Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Unlabelled Data Improves Bayesian Uncertainty Calibration under Covariate Shift
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 |
|---|---|---|---|
| add the transductive variance regularizer using unlabelled target data (ablation) | MC Dropout · Prostate Cancer UK | Error Pred. AUROC: 0.808 $\pm$ 0.011 -> 0.857 $\pm$ 0.009 | Helped |
| add the transductive variance regularizer using unlabelled target data (ablation) | MC Dropout · Prostate Cancer UK | Test Perf. AUROC: 0.836 $\pm$ 0.010 -> 0.861 $\pm$ 0.009 | Helped |
| add the transductive variance regularizer using unlabelled target data (ablation) | MC Dropout · Breast Cancer | Error Pred. AUROC: 0.945 $\pm$ 0.026 -> 0.975 $\pm$ 0.015 | Helped |
| add the transductive variance regularizer using unlabelled target data (ablation) | MC Dropout · Iris | Error Pred. AUROC: 0.657 $\pm$ 0.105 -> 0.877 $\pm$ 0.082 | Helped |
| add the transductive variance regularizer using unlabelled target data (ablation) | MC Dropout · Wine | Error Pred. AUROC: 0.890 $\pm$ 0.055 -> 0.951 $\pm$ 0.034 | Helped |
| vary the monotonic function g used to map posterior variance to source/target confidence (ablation) | Transductive Dropout · Prostate cancer mortality prediction | Uncertainty calibration | No clear effect |
| use transductive regularization with unlabelled target data | MC Dropout · Breast Cancer | Test Perf. AUROC: 0.979 $\pm$ 0.013 -> 0.968 $\pm$ 0.017 | Hurt |
| use transductive regularization with unlabelled target data | MC Dropout · Iris | Test Perf. AUROC: 0.940 $\pm$ 0.045 -> 0.956 $\pm$ 0.045 | Helped |
| use transductive regularization with unlabelled target data | MC Dropout · Wine | Test Perf. AUROC: 1.000 $\pm$ 0.000 -> 1.000 $\pm$ 0.000 | No clear effect |
Closest papers by meaning
- Quantifying Uncertainty in the Presence of Distribution Shifts (NeurIPS 2025 · similarity 0.90)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.86)
- NeuralSurv: Deep Survival Analysis with Bayesian Uncertainty Quantification (NeurIPS 2025 · similarity 0.86)
- Uncertainty Quantification for Deep Regression using Contextualised Normalizing Flows (NeurIPS 2025 · similarity 0.85)
- A Study of Bayesian Neural Network Surrogates for Bayesian Optimization (ICLR 2024 · similarity 0.85)
- Training Bayesian Neural Networks with Sparse Subspace Variational Inference (ICLR 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.