Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Data Augmentation MCMC for Bayesian Inference from Privatized Data
Nianqiao Ju, Jordan Awan, Ruobin Gong, Vinayak Rao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| decrease privacy budget epsilon from 10 to 0.1 | DAMCMC · Log-linear p4 | 90% credible interval coverage: .92 -> .36 | Hurt |
| decrease epsilon from 10 to 0.1 | DAMCMC · Log-linear p1 | 90% credible interval coverage: .92 -> 1 | Mixed |
| decrease epsilon from 10 to 0.3 | DAMCMC · Log-linear p4 | 90% credible interval coverage: .92 -> .59 | Hurt |
| decrease epsilon from 10 to 1 | DAMCMC · Log-linear p4 | 90% credible interval coverage: .92 -> .83 | Hurt |
| decrease epsilon from 10 to 0.1 | DAMCMC · Linear regression beta1 | 90% credible interval coverage: .87 -> .60 | Hurt |
| decrease epsilon from 10 to 1 | DAMCMC · Linear regression beta2 | 90% credible interval coverage: .85 -> .80 | Hurt |
| decrease epsilon from 10 to 3 | DAMCMC · Linear regression beta2 | 90% credible interval coverage: .85 -> .75 | Hurt |
| decrease epsilon from 10 to 0.1 | DAMCMC · Linear regression beta0 | 90% credible interval coverage: .93 -> .99 | No clear effect |
| increase epsilon toward infinity | DAMCMC · Log-linear model posterior mean | posterior mean | Improved |
| decrease epsilon toward zero | DAMCMC · Log-linear model posterior mean | posterior mean | Mixed |
| compare observed acceptance rate with theoretical lower bound exp(-epsilon) | DAMCMC IM proposal · Log-linear and linear regression | acceptance rate | Improved |
Closest papers by meaning
- DP-Fast MH: Private, Fast, and Accurate Metropolis-Hastings for Large-Scale Bayesian Inference (ICML 2023 · similarity 0.89)
- Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy (ICLR 2024 · similarity 0.86)
- On Inherent Privacy of Posterior Sampling: A Unified R\'enyi-Divergence Framework (NeurIPS 2026 · similarity 0.85)
- Privacy Guarantees in Posterior Sampling under Contamination (NeurIPS 2026 · similarity 0.85)
- Differential Privacy Guarantees of Markov Chain Monte Carlo Algorithms (ICML 2025 · similarity 0.84)
- Do You Really Need Public Data? Surrogate Public Data for Differential Privacy on Tabular Data (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.
