Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Uncertainty Estimation via Hyperspherical Confidence Mapping
Eunseo Choi, Ho-Yeon Kim, Jaewon Lee, Taeyong jo, Myungjun lee, Heejin Ahn
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 mixup training to HCM (ablation) | HCM (ResNet-18) · CIFAR-10 OOD detection, average | AUROC (%): 85.04 -> 89.44 | Improved |
| add mixup training to HCM (ablation) | HCM (ResNet-18) · CIFAR-10 near-OOD | AUROC (%): 82.23 -> 87.90 | Improved |
| add mixup training to HCM (ablation) | HCM (ResNet-18) · CIFAR-10 far-OOD | AUROC (%): 86.45 -> 90.12 | Improved |
| add mixup training to MSP baseline (ablation) | MSP · CIFAR-10 OOD detection, average | AUROC (%): -3.61 | Hurt |
| add mixup training to Energy baseline (ablation) | Energy · CIFAR-10 OOD detection, average | AUROC (%): -15.34 | Hurt |
| add mixup training to KNN baseline (ablation) | KNN · CIFAR-10 OOD detection, average | AUROC (%): +2.54 | Improved |
| use HCM instead of Ensembles/MC Dropout/EDL for uncertainty | NYU-v2 depth with noisy input · NYU-v2 depth | Pearson / Spearman: Pearson 0.4919 vs 0.2381 (Ensembles); Spearman 0.5425 vs 0.4684 | Improved |
| use HCM instead of variance-based baselines | NYU-v2 depth with noisy input · NYU-v2 depth | ECE_reg and RMSE: ECE_reg 0.2160 vs 0.0609 (EDL); RMSE 0.1485 vs 0.1189 (MC Dropout) | Hurt |
| use HCM instead of Ensemble/EDL/MC Dropout | Industrial wafer regression · Industrial wafer regression | Pearson correlation: 0.8435 vs 0.3227 (Ensemble) | Improved |
| use HCM instead of Ensemble | Industrial wafer regression · Industrial wafer regression | RMSE: 4.6783 -> 5.4022 | Hurt |
Closest papers by meaning
- Hyperdimensional Uncertainty Quantification for Multimodal Uncertainty Fusion in Autonomous Vehicles Perception (CVPR 2025 · similarity 0.85)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.85)
- Uncertainty Quantification for Physics-Informed Neural Networks with Extended Fiducial Inference (NeurIPS 2025 · similarity 0.85)
- Uncertainty Quantification for Deep Regression using Contextualised Normalizing Flows (NeurIPS 2025 · similarity 0.85)
- Estimating Epistemic and Aleatoric Uncertainty with a Single Model (NeurIPS 2024 · similarity 0.85)
- Learning Polyhedral Conformal Sets for Robust Optimization (NeurIPS 2026 · similarity 0.84)
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.
