Research map / Probabilistic and scientific modeling / Variational and generative sampling
On Testing Conditional Mean Independence for Manifold-Valued Data
Meiling Zeng, Jinhong You, Jicai Liu, Shouxia Wang
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 MMDD test instead of dCov test under variance dependence | conditional mean independence test · S2 DGP1 s2=0.7, alpha=0.1 | rejection rate: 0.987 -> 0.112 | Improved |
| use MMDD test instead of dCov test under variance dependence | conditional mean independence test · W2(R) DGP1 s2=0.7, alpha=0.1 | rejection rate: 0.952 -> 0.110 | Improved |
| increase variance dependence s2 from 0 to 0.7 | MMDD test · S2 DGP1 alpha=0.1 | rejection rate: 0.083 -> 0.112 | No clear effect |
| increase variance dependence s2 from 0 to 0.7 | dCov test · S2 DGP1 alpha=0.1 | rejection rate: 0.096 -> 0.987 | Hurt |
| increase signal strength r2 from 0.1 to 0.5 | MMDD test · S2 DGP2 n=100, alpha=0.1 | power: 0.139 -> 0.950 | Improved |
| increase sample size from 100 to 200 | MMDD test · S2 DGP2 r2=0.3, alpha=0.1 | power: 0.562 -> 0.902 | Improved |
| add heteroscedastic noise (s2=0.5) to linear signal | MMDD test · S2 DGP2 vs DGP3, r2=0.5, n=100 | power: 0.950 -> 0.896 | Hurt |
| add heteroscedastic noise (s2=0.5) to nonlinear signal | MMDD test · S2 DGP4 vs DGP5, r2=0.20, n=200 alpha=0.1 | power: 0.994 -> 0.983 | No clear effect |
| use MMDD instead of dCov for covariate screening PGTM | wind direction data · NOAA wind direction | significance: p < 0.05 vs p > 0.05 | Improved |
Closest papers by meaning
- An Asymptotic Test for Conditional Independence using Analytic Kernel Embeddings (ICML 2022 · similarity 0.87)
- On Distance and Kernel Measures of Conditional Dependence (JMLR 2023 · similarity 0.86)
- On the Hardness of Conditional Independence Testing In Practice (NeurIPS 2025 · similarity 0.85)
- Conditional Diffusion Models are Minimax-Optimal and Manifold-Adaptive for Conditional Distribution Estimation (ICLR 2025 · similarity 0.85)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.84)
- Diffusion Models and the Manifold Hypothesis: Log-Domain Smoothing is Geometry Adaptive (NeurIPS 2025 · 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.
