Research map / Federated learning and privacy / Differential privacy
Private Query Release Assisted by Public Data
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 public data to construct an α-cover, then run PMW on the finite cover and representative domain | PAP query release for finite-VC query classes · Finite-VC query classes | private sample complexity: O((d log(1/α)+log(1/β))^{3/2}√(p log(1/δ))/(ε α^2)) | Helped |
| Use public data to construct an α-cover | PAP query release for finite-VC query classes · Finite-VC query classes | public sample complexity: O((d log(1/α)+log(1/β))/α) | Helped |
| Provide fewer than Ω(1/α²) public samples | PAP query release for decision stumps · Decision stumps on {±1}^p | private sample complexity: n=\tilde\Omega(\sqrt{p}/\alpha) | Hurt |
| Use query release to obtain a PAP learner | Query classes with infinite Littlestone dimension · Infinite-Littlestone-dimension classes | public sample complexity: m=\Omega(1/\alpha) | Hurt |
Closest papers by meaning
- Perturb-and-Project: Differentially Private Similarities and Marginals (ICML 2024 · similarity 0.89)
- Oracle-Efficient Differentially Private Learning with Public Data (NeurIPS 2024 · similarity 0.88)
- Query-Efficient Locally Private Hypothesis Selection via the Scheffe Graph (NeurIPS 2025 · similarity 0.88)
- On Differential Privacy for Adaptively Solving Search Problems via Sketching (ICML 2025 · similarity 0.88)
- Private Prediction via Shrinkage (NeurIPS 2026 · similarity 0.88)
- 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.