Research map / Optimization / Optimizers
On the Convergence of Hamiltonian Monte Carlo with Stochastic Gradients
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 mini-batch SG with SVRG, SAGA, or CVG gradient estimator in HMC (ablation) | stochastic gradient HMC · Multivariate Gaussian synthetic | second-moment estimation error: 0.070 to 0.0017-0.0022 | Helped |
| use CVG instead of SVRG/SAGA as n grows | CVG-HMC · Multivariate Gaussian synthetic | sampling error | No clear effect |
| SG-HMC vs other stochastic HMC variants on Bayesian logistic regression (Covtype) | SG-HMC · Covtype Bayesian logistic regression | sampling error and test negative log-likelihood bias | Hurt |
| SVRG-HMC vs CVG-HMC on small dataset | SVRG-HMC · Covtype Bayesian logistic regression n=500 | sampling error | Helped |
| CVG-HMC on large dataset (n=5000) relative to SVRG-HMC | CVG-HMC · Covtype Bayesian logistic regression n=5000 | test negative log-likelihood | No clear effect |
Closest papers by meaning
- Stochastic Gradient Hamiltonian Monte Carlo Methods with Recursive Variance Reduction (NeurIPS 2019 · similarity 0.94)
- Stochastic Variance-Reduced Hamilton Monte Carlo Methods (ICML 2018 · similarity 0.94)
- A Hybrid Stochastic Gradient Hamiltonian Monte Carlo Method (AAAI 2021 · similarity 0.94)
- A Gradient Based Strategy for Hamiltonian Monte Carlo Hyperparameter Optimization (ICML 2021 · similarity 0.92)
- Hamiltonian Descent Algorithms for Optimization: Accelerated Rates via Randomized Integration Time (NeurIPS 2025 · similarity 0.89)
- Metropolis Adjusted Microcanonical Hamiltonian Monte Carlo (NeurIPS 2025 · similarity 0.87)
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.