Research map / Federated learning and privacy / Differential privacy
Differentially Private 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 |
|---|---|---|---|
| Use DP subsampled MCMC with the proposed GMM correction approximation | 2D Gaussian mixture model posterior sampling · Gaussian mixture model | posterior sample similarity | Helped |
| Replace the ridge-regression correction approximation with a GMM approximation (ablation) | DP subsampled MCMC · Logistic correction distribution, C=2 | approximation error | Helped |
| Increase C above 1 while using the ridge-regression correction approximation (ablation) | DP MCMC correction distribution · Logistic correction distribution | approximation error | Hurt |
| Use smaller minibatches (ablation) | DP subsampled MCMC · Subsampled MCMC acceptance test | acceptance-test approximation error | Hurt |
| Decrease proposal variance (ablation) | DP subsampled MCMC · Gaussian mixture model | proportion of clipped log-likelihood ratios | Helped |
Closest papers by meaning
- Differential Privacy Guarantees of Markov Chain Monte Carlo Algorithms (ICML 2025 · similarity 0.93)
- Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy (ICLR 2024 · similarity 0.93)
- On Inherent Privacy of Posterior Sampling: A Unified R\'enyi-Divergence Framework (NeurIPS 2026 · similarity 0.90)
- Privacy Guarantees in Posterior Sampling under Contamination (NeurIPS 2026 · similarity 0.88)
- Optimal Differentially Private Model Training with Public Data (ICML 2024 · similarity 0.87)
- Beyond the Calibration Point: Mechanism Comparison in Differential Privacy (ICML 2024 · 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.