Research map / Federated learning and privacy / Differential privacy
Poission Subsampled Rényi Differential Privacy
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 tau-term approximation (tau=50) instead of the full alpha-term RDP bound (ablation) | Poisson subsampled RDP bound · Numerical RDP curves (Gaussian, Laplace, randomized response) | closeness to exact RDP bound | Mixed |
| use tight Poisson subsampled RDP bound in moments accountant instead of Abadi et al. Lemma 3 bound | Moments accountant · Privacy loss after k compositions, delta=1e-8 | privacy loss epsilon | Helped |
| moments accountant with tight bound versus optimal (eps,delta)-DP strong composition | Subsampled Gaussian, sigma=5 or 1 · Privacy loss after k compositions | privacy loss epsilon | Mixed |
Closest papers by meaning
- Unified Mechanism-Specific Amplification by Subsampling and Group Privacy Amplification (NeurIPS 2024 · similarity 0.90)
- Differentially Private Stochastic Gradient Descent with Fixed-Size Minibatches: Tighter RDP Guarantees with or without Replacement (NeurIPS 2024 · similarity 0.90)
- Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation (NeurIPS 2026 · similarity 0.89)
- 𝑓-Differential Privacy Filters: Validity and Approximate Solutions (NeurIPS 2026 · similarity 0.88)
- The Gaussian Mixing Mechanism: Renyi Differential Privacy via Gaussian Sketches (NeurIPS 2025 · similarity 0.87)
- Privacy amplification by random allocation (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.