Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Kernel Test of Goodness of Fit
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 |
|---|---|---|---|
| Use the Stein test instead of the B&H test | Goodness-of-fit testing against the normality alternative · Increasing dimensions, n=500 and n=1000 | power: At d=2 and d=5, both tests have power 1; at d=25, B&H has 0.24 vs Stein 0.05 for n=500, and 0.62 vs 0.05 for n=1000 | Mixed |
| Set wild-bootstrap flip probability a_n=0.5 for temporally correlated samples | Stein goodness-of-fit test · Correlated Student's t samples | test calibration | Hurt |
| Set wild-bootstrap flip probability a_n=0.02 | Stein goodness-of-fit test · Correlated Student's t samples | test power | Hurt |
| Thin observations by a factor of 20 and set a_n=0.1 | Stein goodness-of-fit test · Correlated Student's t samples | calibration and power | Helped |
| Use the Stein test rather than model-sample-based MMD for GP model criticism | Gaussian process regression model criticism · solar | p-value: p≈0.1 | No clear effect |
| Increase training data for the infinite-dimensional exponential-family density model | Nonparametric density estimation · Standard Gaussian density; evaluation n=500 | p-value distribution: N=5000 | Helped |
| Use at least 50 random Fourier features | Random Fourier feature density estimator · Standard Gaussian density; evaluation n=500 | null rejection: m=50 | Helped |
Closest papers by meaning
- Sequential Kernel Goodness-of-fit Testing (ICML 2024 · similarity 0.86)
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.85)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.85)
- SD-KDE: Score-Debiased Kernel Density Estimation (NeurIPS 2025 · similarity 0.85)
- Kernel Selection is Model Selection: A Unified Complexity-Penalised Approach for MMD Two-Sample Tests (NeurIPS 2026 · similarity 0.84)
- Sobolev Space Regularised Pre Density Models (ICML 2024 · similarity 0.84)
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.