Research map / Federated learning and privacy / Differential privacy
Revisiting Model-Agnostic Private Learning: Faster Rates and Active Learning
Chong Liu, Yuqing Zhu, Kamalika Chaudhuri, Yu-Xiang Wang
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 active student queries instead of passive queries | PATE-PSQ · real-sim (epsilon=0.5) | test accuracy: 0.6355 -> 0.7389 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · real-sim (epsilon=1.0) | test accuracy: 0.7550 -> 0.8040 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · real-sim (epsilon=2.0) | test accuracy: 0.8025 -> 0.8231 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · a9a (epsilon=2.0) | test accuracy: 0.5176 -> 0.5543 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · mushroom (epsilon=1.0) | test accuracy: 0.7534 -> 0.7727 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · mushroom (epsilon=0.5) | test accuracy: 0.6416 -> 0.6418 | No clear effect |
| use active student queries instead of passive queries | PATE-PSQ · mushroom (epsilon=2.0) | test accuracy: 0.8974 -> 0.8858 | Hurt |
| use active student queries instead of passive queries | PATE-PSQ · mushroom (non-private) | test accuracy: 0.9773 -> 0.9146 | Hurt |
| use active student queries instead of passive queries | PATE-PSQ · a9a (non-private) | test accuracy: 0.5555 -> 0.5461 | No clear effect |
| use active student queries instead of passive queries | PATE-PSQ · real-sim (non-private) | test accuracy: 0.8234 -> 0.8289 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · mushroom (epsilon=1.0) | ex post privacy loss: 1.0 -> 0.9267 | Improved |
| use active student queries instead of passive queries | PATE-PSQ · mushroom | number of labels released: 163 -> 40.1 | Improved |
Closest papers by meaning
- Improved Bounds for Pure Private Agnostic Learning: Item-Level and User-Level Privacy (ICML 2024 · similarity 0.87)
- Oracle-Efficient Differentially Private Learning with Public Data (NeurIPS 2024 · similarity 0.87)
- Certification for Differentially Private Prediction in Gradient-Based Training (ICML 2025 · similarity 0.86)
- Public-data Assisted Private Stochastic Optimization: Power and Limitations (NeurIPS 2024 · similarity 0.86)
- Learning from End User Data with Shuffled Differential Privacy over Kernel Densities (ICLR 2025 · similarity 0.85)
- Computationally Efficient Replicable Learning of Parities and Applications (NeurIPS 2026 · 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.
