Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Deep Evidential Regression
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 |
|---|---|---|---|
| Use evidential regression instead of dropout or ensembles | Regression benchmarks · Boston, Concrete, Energy, Kin8nm, Naval, Power, Protein, Wine, Yacht | NLL | Mixed |
| Use evidential regression instead of dropout or ensembles | Regression benchmarks · Boston, Concrete, Energy, Kin8nm, Naval, Power, Protein, Wine, Yacht | Inference Speed (ms) | Helped |
| Add the error-scaled evidential regularizer (ablation) | Evidential regression · Cubic toy regression, including OOD inputs | Epistemic uncertainty | Helped |
| Set regularization weight to zero (ablation) | Evidential regression · Cubic toy regression, including OOD inputs | Epistemic uncertainty | Hurt |
| Increase regularization weight excessively (ablation) | Evidential regression · Cubic toy regression | Epistemic uncertainty | Hurt |
| Replace the direct evidence regularizer with an epsilon-evidence KL regularizer (ablation) | Evidential regression · Cubic toy regression | Uncertainty calibration | Hurt |
| Use the direct evidence regularizer instead of the epsilon-evidence KL alternative (ablation) | Evidential regression · Cubic toy regression | Epistemic uncertainty calibration | Helped |
| Use evidential regression rather than dropout or ensembling | NYU-Depth-v2 depth estimation · NYU-Depth-v2 test set | Uncertainty calibration error: 0.126; 0.048; 0.033 | Helped |
| Use evidential regression rather than dropout or ensembling | NYU-Depth-v2 depth estimation · NYU-Depth-v2 test set | Inference speed (relative): 1.00 vs 10.20–41.26 for dropout; 1.00 vs 1.94–16.45 for ensembles | Helped |
| Train without OOD examples | Evidential regression · ApolloScape OOD images | Predictive uncertainty | Helped |
Closest papers by meaning
- The Unreasonable Effectiveness of Deep Evidential Regression (AAAI 2023 · similarity 0.93)
- Uncertainty Regularized Evidential Regression (AAAI 2024 · similarity 0.91)
- Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage? (NeurIPS 2024 · similarity 0.90)
- Uncertainty Estimation by Flexible Evidential Deep Learning (NeurIPS 2025 · similarity 0.89)
- The Evidence Contraction Issue in Deep Evidential Regression: Discussion and Solution (AAAI 2024 · similarity 0.88)
- Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods? (ICML 2024 · similarity 0.88)
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.