Research map / Federated learning and privacy / Differential privacy
Easy Differentially Private Linear Regression
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 |
|---|---|---|---|
| increase feature dimension d of synthetic data | TukeyEM · Synthetic | models m needed to pass PTRCheck | Hurt |
| use too few models m | TukeyEM · Synthetic | PTRCheck pass rate | Hurt |
| use too many models m (each trained on few points) | TukeyEM · Beijing | R^2 | Hurt |
| fix m=1000 instead of tuning m | TukeyEM · Real datasets | R^2 | No clear effect |
| use 90th percentile instead of best DPSGD hyperparameters | DPSGD · Real datasets | R^2 | Hurt |
| compare TukeyEM to AdaSSP and DPSGD at (ln 3, 1e-5)-DP | TukeyEM · 8 datasets | median R^2 | Helped |
| remove the outlier row 60422 from Beijing | All DP algorithms · Beijing | R^2 | Helped |
| increase number of models m | TukeyEM · Synthetic | runtime | Hurt |
Closest papers by meaning
- Private Gradient Descent for Linear Regression: Tighter Error Bounds and Instance-Specific Uncertainty Estimation (ICML 2024 · similarity 0.88)
- On the Computational Complexity of Private High-dimensional Model Selection (NeurIPS 2024 · similarity 0.88)
- Better Locally Private Sparse Estimation Given Multiple Samples Per User (ICML 2024 · similarity 0.87)
- Differentially Private High-dimensional Variable Selection via Integer Programming (NeurIPS 2025 · similarity 0.85)
- Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Data (ICML 2024 · similarity 0.85)
- Truthful High Dimensional Sparse Linear Regression (NeurIPS 2024 · similarity 0.85)
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.