Research map / Federated learning and privacy / Differential privacy
Label differential privacy and private training data release
Robert Istvan Busa-Fekete, Andres Munoz medina, Umar Syed, Sergei Vassilvitskii
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 |
|---|---|---|---|
| Replace label-only randomized response with double randomized response (features also resampled from a synthetic conditional GAN) | label-DP training data release · MNIST | label inference accuracy: label inference accuracy drops well below Randomized Response levels | Improved |
| Double randomized response instead of label-only randomized response | label-DP training data release · MNIST | test accuracy: degrades as lambda increases but stays above 95% even when lambda > 1/3 | Mixed |
| Label-only randomized response with fixed public features (static features assumption) | label-DP training data release · Four large binary datasets (kag14, kdd12, kdd10, SUSY) | privacy loss estimate under conditional features: kdd12 instance privacy loss can reach almost 10 versus static-feature epsilon about 7.6 | Hurt |
Closest papers by meaning
- Enhancing Learning with Label Differential Privacy by Vector Approximation (ICLR 2025 · similarity 0.92)
- Certification for Differentially Private Prediction in Gradient-Based Training (ICML 2025 · similarity 0.90)
- Auditing Privacy Mechanisms via Label Inference Attacks (NeurIPS 2024 · similarity 0.89)
- Does Training with Synthetic Data Truly Protect Privacy? (ICLR 2025 · similarity 0.89)
- Differentially Private Sum-Product Networks (ICML 2024 · similarity 0.89)
- Position: Considerations for Differentially Private Learning with Large-Scale Public Pretraining (ICML 2024 · similarity 0.89)
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.
