Research map / Probabilistic and scientific modeling / Variational and generative sampling
Stein Point Markov Chain Monte Carlo
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 global optimization with finite MCMC search using LAST initialization | Stein Points · Gaussian mixture | KSD | Helped |
| replace global optimization with finite MCMC search using INFL initialization | Stein Points · Gaussian mixture | KSD | Helped |
| use RAND initialization instead of LAST or INFL (ablation) | SP-MCMC · Gaussian mixture | KSD | Hurt |
| use INFL initialization to select the chain starting point (ablation) | SP-MCMC · Gaussian mixture | mode-hopping frequency | Helped |
| replace global search with MCMC-based local search | Stein Points · IGARCH | energy distance | Helped |
| replace global search with MCMC-based local search | Stein Points · Goodwin oscillator, d = 4 | KSD | Helped |
| replace global search with MCMC-based local search | Stein Points · Goodwin oscillator, d = 10 | KSD | Helped |
Closest papers by meaning
- Stein Points (ICML 2018 · similarity 0.91)
- Stein $\Pi$-Importance Sampling (NeurIPS 2023 · similarity 0.89)
- Kernel Stein Discrepancy thinning: a theoretical perspective of pathologies and a practical fix with regularization (NeurIPS 2023 · similarity 0.86)
- The Polynomial Stein Discrepancy for Assessing Moment Convergence (ICML 2025 · similarity 0.86)
- AutoStep: Locally adaptive involutive MCMC (ICML 2025 · similarity 0.84)
- Multi-Marginal Couplings for Metropolis--Hastings (NeurIPS 2026 · 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.