Research map / Probabilistic and scientific modeling / Molecules and proteins
Learning $U$-Statistics with Active Inference
Xiaoning Wang, Yuyang Huo, Liuhua Peng, Changliang Zou
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 active sampling with AIPW U-statistic instead of classical uniform IPW | Gini index estimation · Income (ACS) | confidence interval width and effective sample size: roughly 60% labeling-budget reduction | Improved |
| use active sampling instead of classical | Wilcoxon signed-rank U-statistic · Perioperative (VitalDB) | labels needed: about 20% fewer labels | Improved |
| use active sampling instead of uniform AIPW sampling (ablation) | Wilcoxon signed-rank U-statistic · Perioperative (VitalDB) | labels needed: 10% fewer labels | Improved |
| use active sampling instead of classical | Kendall's tau · Political bias | budget for same accuracy: approximately 20% less budget | Improved |
| use active sampling instead of uniform AIPW sampling (ablation) | Kendall's tau · Political bias | budget for same accuracy: about 10% less budget | Improved |
| use active or uniform AIPW sampling instead of classical (ablation) | all three U-statistic tasks · ACS, VitalDB, political bias | empirical coverage of 90% intervals | No clear effect |
| add Hajek-type random normalization to AIPW U-statistic (ablation) | AIPW U-statistic · empirical comparison in appendix | variance | Improved |
| sample by first-order projection residual instead of prediction residual (ablation) | active U-statistic · numerical comparison in appendix | variance | Improved |
Closest papers by meaning
- Active Statistical Inference (ICML 2024 · similarity 0.88)
- Robust Sampling for Active Statistical Inference (NeurIPS 2025 · similarity 0.87)
- Prediction-Powered Active Testing (NeurIPS 2026 · similarity 0.84)
- Uncertainty Herding: One Active Learning Method for All Label Budgets (ICLR 2025 · similarity 0.83)
- Sample-Optimal Agnostic Boosting with Unlabeled Data (ICML 2025 · similarity 0.83)
- Post-ADC Inference: Valid Inference After Active Data Collection (NeurIPS 2026 · similarity 0.83)
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.
