Research map / Probabilistic and scientific modeling / Variational and generative sampling
Random Feature Stein Discrepancies
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 |
|---|---|---|---|
| Replace quadratic-time KSD with RPhiSD (M=10 features) | Stein discrepancy · Compute time, D=10, N 500-5000 | runtime | Helped |
| Replace quadratic-time IMQ KSD with RPhiSD for step-size selection | SGLD hyperparameter selection · Bimodal GMM posterior, N=1000 | selected step size | No clear effect |
| Use L1 IMQ instead of L2 SechExp | RPhiSD · Importance-sample efficiency | M needed for small variance | Helped |
| Use L1 IMQ RPhiSD instead of FSSD and RFF tests | goodness-of-fit testing · Gaussian vs Laplace / t / RBM | test power | Helped |
| Use L2 SechExp RPhiSD instead of FSSD/RFF | goodness-of-fit testing · Gaussian null alternatives | test power | No clear effect |
| Use L1 IMQ RPhiSD in place of quadratic-time KSD | goodness-of-fit testing · Gaussian and RBM alternatives | test power | No clear effect |
| Use r=1 (L1) instead of r=2 feature norm (ablation) | RPhiSD · importance sample complexity | required M | Helped |
Closest papers by meaning
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.92)
- Measuring Sample Quality with Kernels (ICML 2017 · similarity 0.92)
- Gradient-Free Kernel Stein Discrepancy (NeurIPS 2023 · similarity 0.91)
- Stochastic Stein Discrepancies (NeurIPS 2020 · similarity 0.91)
- Revealing Distribution Discrepancy by Sampling Transfer in Unlabeled Data (NeurIPS 2024 · similarity 0.85)
- Anchor-based Maximum Discrepancy for Relative Similarity Testing (NeurIPS 2025 · similarity 0.85)
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.