Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Kernelized Stein Discrepancy for Goodness-of-fit Tests
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 U-statistic KSD linear estimator (KSD-Linear) instead of U-statistic KSD-U (ablation) | goodness-of-fit test on 1D Gaussian mixture and RBM · 1D Gaussian mixture; Gaussian-Bernoulli RBM | error rate | Hurt |
| Replace exact-sample MMD (MMD-MC) with MCMC-approximated sample (MMD-MCMC) | two-sample MMD test on RBM · Gaussian-Bernoulli RBM | error rate | Hurt |
| Increase exact Monte Carlo sample size n' for MMD-MC from 100 to 1000 (ablation) | MMD two-sample test on RBM · Gaussian-Bernoulli RBM | error rate | Helped |
| KSD-U goodness-of-fit test compared with classical tests on 1D Gaussian mixture | goodness-of-fit testing · 1D Gaussian mixture | error rate | Helped |
| Use AIS-approximated likelihood ratio test instead of KSD-U | RBM goodness-of-fit · Gaussian-Bernoulli RBM | discrimination of p=q vs p!=q | Hurt |
Closest papers by meaning
- Goodness-of-Fit Testing for Discrete Distributions via Stein Discrepancy (ICML 2018 · similarity 0.96)
- A Kernel Stein Test of Goodness of Fit for Sequential Models (ICML 2023 · similarity 0.94)
- A Kernel Test of Goodness of Fit (ICML 2016 · similarity 0.92)
- Using Perturbation to Improve Goodness-of-Fit Tests based on Kernelized Stein Discrepancy (ICML 2023 · similarity 0.92)
- Random Feature Stein Discrepancies (NeurIPS 2018 · similarity 0.90)
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.89)
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.