Research map / Federated learning and privacy / Differential privacy
On the Differential Privacy of Bayesian Inference
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 |
|---|---|---|---|
| add Fourier-domain perturbation (stealth, consistent updates) in place of direct Laplace noise on posterior updates (ablation) | private Bayesian naive Bayes · synthetic naive Bayes, 50 training examples | classification accuracy | Hurt |
| use posterior sampler instead of Laplace mechanisms at larger epsilon | private Bayesian naive Bayes · synthetic naive Bayes, 50 training examples | classification accuracy | Helped |
| use posterior sampler at small epsilon (large trimming probability) | private Bayesian naive Bayes · synthetic naive Bayes, 50 training examples | classification accuracy | Hurt |
| add differential privacy noise to naive Bayes posterior | Bayesian naive Bayes · synthetic naive Bayes, 50 training examples | classification accuracy | Hurt |
| make the prior more concentrated (larger precision b) | private Bayesian linear regression · IPUMS census, annual income | MSE | Hurt |
| private posterior sampling instead of non-private linear regression | Bayesian linear regression · IPUMS census, annual income | MSE | Hurt |
Closest papers by meaning
- Differentially Private Bayesian Inference for Exponential Families (NeurIPS 2018 · similarity 0.91)
- Privacy Guarantees in Posterior Sampling under Contamination (NeurIPS 2026 · similarity 0.91)
- On Inherent Privacy of Posterior Sampling: A Unified R\'enyi-Divergence Framework (NeurIPS 2026 · similarity 0.90)
- Locally Private Bayesian Inference for Count Models (ICML 2019 · similarity 0.90)
- Differential Privacy Guarantees of Markov Chain Monte Carlo Algorithms (ICML 2025 · similarity 0.88)
- Tractable MCMC for Private Learning with Pure and Gaussian Differential Privacy (ICLR 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.