Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Centroid Approximation for Bootstrap: Improving Particle Quality at Inference
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 |
|---|---|---|---|
| Replace i.i.d. bootstrap particles with jointly optimized centroids | Bootstrap confidence intervals · Linear-model CI, alpha=0.9, Normal/Percentile/Pivotal, m=20 and m=50 | coverage error: Normal m=20: 0.029±0.010 -> 0.027±0.010; Percentile m=20: 0.101±0.013 -> 0.081±0.012; Pivotal m=50: 0.045±0.011 -> 0.013 | Helped |
| Replace i.i.d. bootstrap particles with jointly optimized centroids | Contextual-bandit bootstrap · Mushroom, Statlog, Financial, m=3 and m=10 | cumulative reward: Mushroom m=3: 3282.1±72.8 -> 3702.7±89.8; Statlog m=10: 1877.0±2.7 -> 1892.6±2.8; Financial m=10: 2281.4±56.6 -> 2325.5± | Helped |
| Add the low-assignment common-gradient fallback (ablation) | Centroid contextual-bandit method · Mushroom, m=3, 4, 5, 10 | cumulative reward: m=3: 3480.0±120 -> 3702.7±89.8; m=4: 3461.9±126 -> 3723.1±78.7; m=5: 3586.5±64.5 -> 3799.6±84.2; m=10: 3785.0±59.1 -> 37 | Helped |
| Retain bootstrap uncertainty instead of always using the common-data gradient (ablation) | Centroid contextual-bandit method · Mushroom, m=3, 4, 5, 10 | cumulative reward: m=3: 3467.7±115 -> 3702.7±89.8; m=4: 3600.0±69.3 -> 3723.1±78.7; m=5: 3647.3±64.5 -> 3799.6±84.2; m=10: 3742.7±86.8 -> 3 | Helped |
Closest papers by meaning
- Bootstrap in High Dimension with Low Computation (ICML 2023 · similarity 0.89)
- Orthogonal Bootstrap: Efficient Simulation of Input Uncertainty (ICML 2024 · similarity 0.87)
- Simultaneous Inference for Massive Data: Distributed Bootstrap (ICML 2020 · similarity 0.85)
- Gaussian Approximation and Multiplier Bootstrap for Polyak-Ruppert Averaged Linear Stochastic Approximation with Applications to TD Learning (NeurIPS 2024 · similarity 0.81)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.81)
- Statistical inference for Linear Stochastic Approximation with Markovian Noise (NeurIPS 2025 · similarity 0.80)
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.