Research map / Probabilistic and scientific modeling / Variational and generative sampling
Evaluating the Implicit Midpoint Integrator for Riemannian Hamiltonian 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 |
|---|---|---|---|
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Banana-shaped distribution, step size 0.1, 5 steps | Mean ESS / Sec.: 3.36 ± 0.13 -> 9.03 ± 0.28 | Helped |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Neal's funnel, step size 0.5, 20 steps | Min. ESS / Sec.: 0.74 ± 0.06 -> 3.58 ± 0.13 | Helped |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Stochastic volatility model | Mean ESS: 297.82 ± 9.2 -> 309.10 ± 10.5 | Helped |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Fitzhugh-Nagumo, step size 1.0, 1 step | Mean ESS / Sec.: 0.21 ± 0.01 -> 0.06 ± 0.00 | Hurt |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Fitzhugh-Nagumo, step size 1.0, 5 steps | Mean ESS / Sec.: 0.03 ± 0.00 -> 0.05 ± 0.01 | Helped |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Breast Cancer logistic regression, step size 1.0, 10 steps | Mean ESS / Sec.: 1.48 ± 0.16 -> 35.43 ± 2.34 | Helped |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Breast Cancer logistic regression, step size 0.1, 10 steps | Mean ESS / Sec.: 29.85 ± 0.19 -> 14.42 ± 0.16 | Hurt |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Diabetes logistic regression, step size 1.0, 10 steps | Mean ESS / Sec.: 30.17 ± 0.28 -> 38.39 ± 0.08 | Helped |
| Use implicit midpoint instead of generalized leapfrog | RMHMC · Heart disease logistic regression, step size 1.0, 10 steps | Mean ESS / Sec.: 0.51 ± 0.11 -> 5.72 ± 0.20 | Helped |
| Use the midpoint-then-explicit-Euler implementation instead of direct implicit midpoint (ablation) | I.M. · Banana-shaped distribution, step size 0.1, 5 steps | Mean ESS / Sec.: 8.65 ± 0.31 -> 9.03 ± 0.28 | No clear effect |
| Cache repeated calculations in generalized leapfrog (ablation) | G.L.F. · Banana-shaped distribution, step size 0.1, 5 steps | Mean ESS / Sec.: 1.21 ± 0.04 -> 3.36 ± 0.13 | Helped |
Closest papers by meaning
- Robust Monte Carlo Sampling using Riemannian Nosé-Poincaré Hamiltonian Dynamics (ICML 2016 · similarity 0.89)
- Practical Hamiltonian Monte Carlo on Riemannian Manifolds via Relativity Theory (ICML 2024 · similarity 0.89)
- Unbiased constrained sampling with Self-Concordant Barrier Hamiltonian Monte Carlo (NeurIPS 2023 · similarity 0.88)
- Towards Unifying Hamiltonian Monte Carlo and Slice Sampling (NeurIPS 2016 · similarity 0.87)
- Metropolis Adjusted Microcanonical Hamiltonian Monte Carlo (NeurIPS 2025 · similarity 0.85)
- Hamiltonian Descent Algorithms for Optimization: Accelerated Rates via Randomized Integration Time (NeurIPS 2025 · 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.