Research map / Federated learning and privacy / Differential privacy
Learning from Mixtures of Private and Public Populations
Raef Bassily, Shay Moran, Anupama Nandi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 examples to construct a finite hypothesis class, then select from it with the exponential mechanism | Halfspaces in R^d under label-determined privacy · Realizable learning | sample complexity: O\left(\frac{d^2 \log(d/\epsilon\alpha) + \log(1/\beta)}{~\epsilon~\alpha}\right) | Improved |
| Use public examples to construct a finite hypothesis class, then select from it with the exponential mechanism | Halfspaces in R^d under label-determined privacy · Agnostic learning | sample complexity: O\left(\left(d^2 \log\left(\frac{d}{\epsilon\alpha}\right) + \log(\frac{1}{\beta})\right)\max\left(\frac{1}{\alpha^2},\f | Improved |
| Allow privacy status to be correlated with the target label instead of treating every example as private | Learning halfspaces in R^d · Label-determined private-public mixture | learnability | Improved |
Closest papers by meaning
- Limits of Private Learning with Access to Public Data (NeurIPS 2019 · similarity 0.88)
- Oracle-Efficient Differentially Private Learning with Public Data (NeurIPS 2024 · similarity 0.85)
- It's My Data Too: Private ML for Datasets with Multi-User Training Examples (ICML 2025 · similarity 0.85)
- Differentially Private Model Merging (NeurIPS 2026 · similarity 0.85)
- Differential Privacy Under Class Imbalance: Methods and Empirical Insights (ICML 2025 · similarity 0.85)
- Certification for Differentially Private Prediction in Gradient-Based Training (ICML 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.
