Research map / Federated learning and privacy / Differential privacy
Differentially Private Sampling from Distributions
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 private sampling rather than privately learning a distribution | Distributions on [k] · Distributions on [k] | sample complexity: Θ\paren{\dfrac{k}{\alpha \varepsilon}} vs. Θ\paren{\dfrac{k}{\alpha^2} + \dfrac{k}{\alpha \varepsilon}} | Helped |
| Use private sampling rather than privately learning a distribution | Product distributions on {0,1}^d · Product distributions on {0,1}^d | sample complexity: \tilde \Theta\paren{\dfrac{d }{\alpha \varepsilon}} vs. \tilde \Theta\paren{\dfrac{d }{\alpha \varepsilon} + \dfrac{d}{\ | Helped |
| Restrict product-distribution biases to [1/3, 2/3] | Private sampling for product distributions · Product distributions on {0,1}^d | sample complexity: \tilde \Theta\paren{\dfrac{d }{\alpha \varepsilon}} -> \tilde O \paren{\dfrac{\sqrt{d}}{\varepsilon} + \log\dfrac{d}{\al | Helped |
| Use private sampling instead of nonprivate learning for bounded-bias product distributions | Product distributions with biases in [1/3, 2/3] · Product distributions on {0,1}^d, biases in [1/3, 2/3] | sample complexity: \Theta\paren{ \dfrac{d}{\alpha^2}} vs. \tilde O \paren{\dfrac{\sqrt{d}}{\varepsilon} + \log\dfrac{d}{\alpha}} | Helped |
Closest papers by meaning
- On Differentially Private Sampling from Gaussian and Product Distributions (NeurIPS 2023 · similarity 0.92)
- Exactly Minimax-Optimal Locally Differentially Private Sampling (NeurIPS 2024 · similarity 0.90)
- Locally Optimal Private Sampling: Beyond the Global Minimax (NeurIPS 2025 · similarity 0.89)
- Sample-Efficient Private Learning of Mixtures of Gaussians (NeurIPS 2024 · similarity 0.88)
- Privacy Guarantees in Posterior Sampling under Contamination (NeurIPS 2026 · similarity 0.87)
- Public-data Assisted Private Stochastic Optimization: Power and Limitations (NeurIPS 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.