Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
What is Flagged in Uncertainty Quantification? Latent Density Models for Uncertainty Categorization
Hao Sun, Boris van Breugel, Jonathan Crabbé, Nabeel Seedat, Mihaela van der Schaar
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 Deep Ensemble as the UQ method instead of MC-Dropout | DAUC categorization · Dirty-MNIST, top 5% uncertain | share of flagged examples that are OOD | Improved |
| use MC-Dropout instead of Deep Ensemble | UQ method · Dirty-MNIST | OOD flagging | Hurt |
| use Deep Ensemble rather than BNN or MCD | UQ method · Dirty-CIFAR-10 | Bnd and IDM flagging | Improved |
| use BNN instead of Deep Ensemble | UQ method · Dirty-CIFAR-10 | Bnd and IDM flagging | Hurt |
| vary Bnd and IDM thresholds away from 0.8 (ablation) | DAUC · Dirty-MNIST | categorization conclusions | No clear effect |
| change kernel from Gaussian to Tophat or Exponential (ablation) | DAUC · Dirty-MNIST | score correlation: all above 0.86 | No clear effect |
| replace DAUC OOD scoring with Outlier-AE (ablation) | OOD detector · Dirty-MNIST | F1-score: 1.0000 -> 1.0000 | No clear effect |
| replace DAUC OOD scoring with IForest (ablation) | OOD detector · Dirty-MNIST | F1-score: 1.0000 -> 0.9999 | No clear effect |
| retrain on training data filtered to resemble B&I test examples (increase discarded proportion q) | linear models and ResNet-18 · Covtype, Digits, Spam, DMNIST | accuracy on B&I class | Improved |
| discard too much training data (large q) (ablation) | filtered retraining · Covtype, Digits, Spam, DMNIST | accuracy on B&I class | Hurt |
| filtered retraining on Bnd class | linear models · Covtype, Digits, Spam | accuracy on Bnd class | Improved |
| filtered retraining on IDM class | linear models · Covtype, Digits, Spam | accuracy on IDM class | Improved |
Closest papers by meaning
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.87)
- Second-Order Uncertainty Quantification: A Distance-Based Approach (ICML 2024 · similarity 0.87)
- Efficient Analytic Uncertainty Quantification for Multimodal Regression (NeurIPS 2026 · similarity 0.86)
- Epistemic Uncertainty Quantification For Pre-Trained Neural Networks (CVPR 2024 · similarity 0.86)
- Linear Uncertainty Quantification of Graphical Model Inference (NeurIPS 2024 · similarity 0.86)
- Bernstein–von Mises for Adaptively Collected Data (NeurIPS 2025 · 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.
