Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Improving Evidential Deep Learning via Multi-Task Learning
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 Lipschitz modified MSE to ENet's NLL objective | ENet · UCI regression benchmarks | RMSE: Boston: 3.06(0.16) -> 3.04(0.21); Concrete: 5.85(0.15) -> 5.60(0.17); Protein: 4.64(0.03) -> 4.73(0.07) | Mixed |
| Add Lipschitz modified MSE to ENet's NLL objective | ENet · UCI regression benchmarks | NLL: Boston: 2.35(0.06) -> 2.31(0.04); Kin8nm: -1.24(0.01) -> -1.19(0.01) | Mixed |
| Add Lipschitz modified MSE to ENet | ENet · Davis | ECE: 0.184(0.02) -> 0.156(0.03) | Helped |
| Add Lipschitz modified MSE to ENet | ENet · Kiba | MSE: 0.190(0.00) -> 0.181(0.00) | Helped |
| Add Lipschitz modified MSE to ENet | ENet · BindingDB | ECE: 0.018(0.01) -> 0.015(0.00) | Helped |
| Add simple MSE as an additional loss instead of Lipschitz modified MSE (ablation) | ENet · Kiba | MSE: 0.190(0.00) -> 0.176(0.00) | Helped |
| Add simple MSE as an additional loss instead of Lipschitz modified MSE (ablation) | ENet · Kiba | ECE: 0.077(0.03) -> 0.081(0.02) | Hurt |
| Add simple MSE as an additional loss instead of Lipschitz modified MSE (ablation) | ENet · BindingDB | ECE: 0.018(0.01) -> 0.027(0.01) | Hurt |
| Replace simple MSE with Lipschitz modified MSE (ablation) | ENet with an additional MSE task · DTA gradient cosine similarity | gradient cosine similarity | Helped |
| Reduce the weight of simple MSE to 0.1 rather than using Lipschitz modification (ablation) | ENet with an additional MSE task · DTA gradient cosine similarity | gradient cosine similarity | No clear effect |
Closest papers by meaning
- Uncertainty Regularized Evidential Regression (AAAI 2024 · similarity 0.89)
- Are Uncertainty Quantification Capabilities of Evidential Deep Learning a Mirage? (NeurIPS 2024 · similarity 0.89)
- The Evidence Contraction Issue in Deep Evidential Regression: Discussion and Solution (AAAI 2024 · similarity 0.87)
- Uncertainty Estimation by Flexible Evidential Deep Learning (NeurIPS 2025 · similarity 0.87)
- Calibrating LLMs with Information-Theoretic Evidential Deep Learning (ICLR 2025 · similarity 0.86)
- Uncertainty Estimation by Density Aware Evidential Deep Learning (ICML 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.