Research map / Federated learning and privacy / Differential privacy
Leveraging Public Data for Practical Private Query Release
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 PA-10, OH-18, or NY-18 as public data to define the support and prior | PMW^Pub · ACS (reduced) PA-18, 5-way marginals | max error | Helped |
| Use public data to define the support and prior | PMW^Pub · ACS PA-18 and GA-18, 3-way marginals | max error | Helped |
| Use CA-18 public data, whose support has high best-mixture error for the workload (ablation) | PMW^Pub · ACS PA-18 and GA-18 | max error | Hurt |
| Use unbiased public data (Δ = 0) (ablation) | PMW^Pub · ADULT, 3-way marginals | max error | Helped |
| Use public data with low sex-sampling bias (|Δ| ≤ 0.2) (ablation) | PMW^Pub · ADULT, 3-way marginals | max error | Helped |
| Use extremely sex-biased public data (Δ ∈ { 0.45, 0.65 }) (ablation) | PMW^Pub · ADULT, 3-way marginals | max error | Hurt |
| Use only 1% of the PA-10 or OH-18 public dataset (ablation) | PMW^Pub · ACS PA-18, 3-way marginals | max error | Helped |
| Replace MWEM's full domain with the public dataset's support | PMW^Pub · ACS (reduced) PA-18 | per-iteration run-time: 0.919 -> 0.185 seconds | Helped |
Closest papers by meaning
- Do You Really Need Public Data? Surrogate Public Data for Differential Privacy on Tabular Data (NeurIPS 2025 · similarity 0.87)
- Private Query Release Assisted by Public Data (ICML 2020 · similarity 0.87)
- A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy (AAAI 2024 · similarity 0.87)
- RAPID: Retrieval Augmented Training of Differentially Private Diffusion Models (ICLR 2025 · similarity 0.85)
- Privacy-Preserving Data Release Leveraging Optimal Transport and Particle Gradient Descent (ICML 2024 · similarity 0.84)
- Fairness Issues and Mitigations in (Differentially Private) Socio-Demographic Data Processes (AAAI 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.