Research map / Probabilistic and scientific modeling / Variational and generative sampling
Continuously Tempered PDMP samplers
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 |
|---|---|---|---|
| Add continuous tempering with a point mass at β=1 | Zig-Zag · Gaussian mixture | RMSE of E[X1]: 2.557 -> 0.304 | Helped |
| Add continuous tempering with a point mass at β=1 | Zig-Zag · Gaussian mixture | RMSE of E[X1²]: 30.577 -> 3.216 | Helped |
| Allow excursions to lower temperatures | Zig-Zag · Gaussian mixture | thinning efficiency: 0.057 -> 0.139 | Helped |
| Use importance sampling without a point mass at β=1 | Continuously tempered Zig-Zag · Gaussian mixture | RMSE of E[X1]: 0.294 -> 0.483 | Hurt |
| Set α to 0.8 rather than 0.3 (ablation) | Zig-Zag CT · Gaussian mixture | RMSE of E[X1]: 0.304 -> 0.650 | Hurt |
| Set α to 0.7 rather than 0.3 (ablation) | Zig-Zag CT · Gaussian mixture | RMSE of E[X1]: 0.304 -> 0.399 | Hurt |
| Set α to 0.5 rather than 0.3 (ablation) | Zig-Zag CT · Gaussian mixture | RMSE of E[X1]: 0.304 -> 0.329 | Hurt |
| Set α to 0.1 rather than 0.3 (ablation) | Zig-Zag CT · Gaussian mixture | RMSE of E[X1]: 0.304 -> 0.349 | Hurt |
| Add continuous tempering with a point mass at β=1 | Transdimensional Zig-Zag · Transdimensional spike-slab, m=4 | mean absolute error of E[X1]: 1.998 -> 0.214 | Helped |
| Add continuous tempering with a point mass at β=1 | Transdimensional Zig-Zag · Transdimensional spike-slab, m=4 | mean absolute error of P(|X1| > 0): 0.500 -> 0.055 | Helped |
| Add continuous tempering with a point mass at β=1 | Transdimensional Zig-Zag · Transdimensional spike-slab, m=0 | mean absolute error of E[X1]: 0.005 -> 0.007 | Hurt |
| Add continuous tempering with a point mass at β=1 | Zig-Zag · Boltzmann machine relaxation | average RMSE of E[Xk]: 1.304 -> 0.296 | Helped |
Closest papers by meaning
- Restricted Spectral Gap Decomposition for Simulated Tempering Targeting Mixture Distributions (NeurIPS 2025 · similarity 0.89)
- Piecewise deterministic generative models (NeurIPS 2024 · similarity 0.88)
- Provable Convergence and Limitations of Geometric Tempering for Langevin Dynamics (ICLR 2025 · similarity 0.87)
- Progressive Tempering Sampler with Diffusion (ICML 2025 · similarity 0.86)
- Wasserstein Gradient Flows and Forward-Only Diffusion Are Not Enough for Multimodal Sampling (NeurIPS 2026 · similarity 0.86)
- Chain of Log-Concave Markov Chains (ICLR 2024 · similarity 0.86)
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.