Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
The Implicit Delta Method
Nathan Kallus, James McInerney
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 FDIDM instead of bootstrap to estimate uncertainty | Neural network classifier · Vehicle | runtime: 806 -> 39 seconds | Improved |
| Use FDIDM instead of bootstrap to estimate uncertainty | Neural network classifier · Waveform | runtime: 2,334 -> 129 seconds | Improved |
| Use FDIDM instead of bootstrap to estimate uncertainty | Neural network classifier · Satellite | runtime: 3,192 -> 111 seconds | Improved |
| Use FDIDM instead of bootstrap to estimate uncertainty | Neural network classifier · MNIST | runtime: 7,164 -> 303 seconds | Improved |
| Set finite-difference width to λ=0.512 rather than the delta-method estimate (ablation) | FDIDM · Quadratic synthetic task | root mean squared error of variance estimate | Improved |
| Use very small finite-difference widths λ<0.01 (ablation) | FDIDM · Quadratic synthetic task | root mean squared error of variance estimate | Hurt |
| Increase KL down-weighting T and introduce more reconstruction-quality variance (ablation) | Variational autoencoder · MNIST | reconstruction quality | Improved |
Closest papers by meaning
- General Uncertainty Estimation with Delta Variances (AAAI 2025 · similarity 0.91)
- Epistemic Uncertainty Quantification For Pre-Trained Neural Networks (CVPR 2024 · similarity 0.88)
- Quantifying Uncertainty in the Presence of Distribution Shifts (NeurIPS 2025 · similarity 0.86)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.86)
- One Step Closer to Unbiased Aleatoric Uncertainty Estimation (AAAI 2024 · similarity 0.86)
- Linear Uncertainty Quantification of Graphical Model Inference (NeurIPS 2024 · similarity 0.86)
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.
