Research map / Federated learning and privacy / Differential privacy
Private Rate-Constrained Optimization with Applications to Fair Learning
Mohammad Yaghini, Tudor Cebere, Michael Menart, Aurélien Bellet, Nicolas Papernot
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 RaCO-DP instead of prior DP fairness baselines (DP-FERMI etc.) | RaCO-DP · Adult, Credit-Card, Parkinsons | accuracy at fixed disparity | Improved |
| use RaCO-DP instead of DP-FERMI for training time per step | RaCO-DP · Adult | time per step: three orders of magnitude faster | Improved |
| move from logistic regression to ResNet16 on CelebA (privacy gap) | RaCO-DP · CelebA | accuracy: non-private 95% vs RaCO-DP 90% at eps=1 | Hurt |
| lower clipping norm below 12.5 under strict FNR=0 (ablation) | RaCO-DP · Adult | constraint satisfaction: norms below 12.5 prevent meeting the constraint | Hurt |
| use hard constraints instead of soft constraints in dual update (ablation) | RaCO-DP · Adult | utility | No clear effect |
| use tempered-sigmoid temperature tau=1 (ablation) | RaCO-DP · Adult | hard constraint satisfaction | Improved |
Closest papers by meaning
- Faster Algorithms for User-Level Private Stochastic Convex Optimization (NeurIPS 2024 · similarity 0.89)
- Online Sensitivity Optimization in Differentially Private Learning (AAAI 2024 · similarity 0.89)
- Differential Private Stochastic Optimization with Heavy-tailed Data: Towards Optimal Rates (AAAI 2025 · similarity 0.88)
- Improved Rates of Differentially Private Nonconvex-Strongly-Concave Minimax Optimization (AAAI 2025 · similarity 0.88)
- Towards hyperparameter-free optimization with differential privacy (ICLR 2025 · similarity 0.88)
- Differentially Private Optimization with Sparse Gradients (NeurIPS 2024 · similarity 0.88)
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.
