Research map / Learning theory and algorithms / Sample complexity
Accelerating Hamiltonian Monte Carlo via Chebyshev Integration Time
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 constant integration time with Chebyshev integration time | Ideal HMC · Gaussian, covariance diag(1, 100) | Mean ESS: $7692.00382 \pm 207.19628$ -> $10399.00811 \pm 347.25021$ | Helped |
| Permute the Chebyshev integration times (ablation) | Ideal HMC · Gaussian, covariance diag(1, 100) | Mean ESS: $10197.09964 \pm 276.94894$ -> $10399.00811 \pm 347.25021$ | Helped |
| Replace constant integration time with Chebyshev integration time | HMC · Gaussian | Mean ESS: $1852.79 \pm 132.95$ -> $4948.46 \pm 144.03$ | Helped |
| Replace constant integration time with Chebyshev integration time | HMC · Mixture of two Gaussians | Mean ESS: $632.70 \pm 22.78$ -> $2064.42 \pm 67.44$ | Helped |
| Replace constant integration time with Chebyshev integration time | Bayesian logistic regression HMC · Heart | Mean ESS: $242.44 \pm 14.61$ -> $1424.21 \pm 54.03$ | Helped |
| Replace constant integration time with Chebyshev integration time | Bayesian logistic regression HMC · Breast Cancer | Mean ESS: $99.48 \pm 10.10$ -> $886.24 \pm 38.92$ | Helped |
| Replace constant integration time with Chebyshev integration time | Bayesian logistic regression HMC · Diabetes | Mean ESS: $57.11 \pm 9.52$ -> $546.80 \pm 37.40$ | Helped |
| Replace constant integration time with Chebyshev integration time | HMC · Hard distribution $f_{0.05}$ | Mean ESS: $2796.41 \pm 56.89$ -> $6457.21 \pm 110.05$ | Helped |
Closest papers by meaning
- On the Convergence of Hamiltonian Monte Carlo with Stochastic Gradients (ICML 2021 · similarity 0.89)
- Stochastic Variance-Reduced Hamilton Monte Carlo Methods (ICML 2018 · similarity 0.89)
- Entropy-based adaptive Hamiltonian Monte Carlo (NeurIPS 2021 · similarity 0.88)
- Hamiltonian Descent Algorithms for Optimization: Accelerated Rates via Randomized Integration Time (NeurIPS 2025 · similarity 0.87)
- Practical Hamiltonian Monte Carlo on Riemannian Manifolds via Relativity Theory (ICML 2024 · similarity 0.85)
- Metropolis Adjusted Microcanonical Hamiltonian Monte Carlo (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.