Research map / Learning theory and algorithms / Kernels and regression theory
Multiple-Splitting Projection Test for High-Dimensional Mean Vectors
Wanjun Liu, Xiufan Yu, Runze Li
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 multiple splits (MPT) instead of a single split (SPT) | projection test · bone density, delta=0.18 | reject null at 0.05: 0.246 -> reject | Improved |
| use multiple splits (MPT) instead of a single split (SPT) | projection test · simulation, normal and t6 samples | power | Improved |
| use MPT instead of Cauchy combination | p-value combination · CS covariance simulation, n=40 | power | Improved |
| use MPT instead of median, average, Z-average combination | p-value combination · AR covariance simulation, n=40 | power | Improved |
| use MPT instead of CLX test | extreme-type test · simulation, AR covariance, size under H0 | type I error: as large as 0.20 | Improved |
| use MPT instead of CQ test | quadratic-form test · CS covariance simulation, power vs correlation | power | Improved |
| use MPT instead of CQ/CLX/average at weak signal | MPT · bone density, delta=0.3 | reject null at 0.05: 0.081 -> reject | Improved |
| use heavy-tailed t6 data instead of normal data | MPT · simulation n=40 | size and power | No clear effect |
| increase sample size from n=40 to n=100 | MPT · simulation | size and power | No clear effect |
| exploit exchangeability of p-values using worst-case rho critical value | MPT · simulation size | type I error | Hurt |
Closest papers by meaning
- High-Dimensional Conditional Independence Testing via Random Projection Aggregation (NeurIPS 2026 · similarity 0.86)
- Randomized tests for high-dimensional regression: A more efficient and powerful solution (NeurIPS 2020 · similarity 0.85)
- Two Sample Testing in High Dimension via Maximum Mean Discrepancy (JMLR 2023 · similarity 0.83)
- Optimal Algorithms for Augmented Testing of Discrete Distributions (NeurIPS 2024 · similarity 0.81)
- Debiasing Random Oblique Projections for Subsampled OLS and Fast CUR in High Dimensions (NeurIPS 2026 · similarity 0.80)
- Testing Calibration in Nearly-Linear Time (NeurIPS 2024 · 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.
