Research map / Federated learning and privacy / Differential privacy
Censoring with Plausible Deniability: Asymmetric Local Privacy for Multi-Category CDF Estimation
Qirui Hu, Yi Liu
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 |
|---|---|---|---|
| censor the categorical label for sensitive outputs (ACRR) instead of utility-optimized randomized response | ULDP randomized response · Truthful report probability, sensitive outputs (K=4, epsilon=1) | truthful report probability (%): 47% -> 100% | Improved |
| use ACRR instead of ULDP randomized response for safe inputs | ULDP randomized response · Truthful report probability, safe inputs (K=4, epsilon=1) | truthful report probability (%): 30% -> 63% | Improved |
| use ACRR instead of standard LDP randomized response | LDP randomized response · Truthful report probability (K=4, epsilon=1) | truthful report probability (%): 28% -> 63% | Improved |
| divide data into M = 4 subsets and average the estimators (ablation) | ICM CDF estimator · Estimation accuracy and runtime | estimation error | Improved |
| cap total CDF at 1 and freeze sub-CDFs (ablation) | ACRR CDF estimator · Estimation error | estimation error | Improved |
Closest papers by meaning
- Tuning-free Estimation and Inference of Cumulative Distribution Function under Local Differential Privacy (ICML 2024 · similarity 0.90)
- Instance-Specific Asymmetric Sensitivity in Differential Privacy (NeurIPS 2024 · similarity 0.89)
- A Generalized Shuffle Framework for Privacy Amplification: Strengthening Privacy Guarantees and Enhancing Utility (AAAI 2024 · similarity 0.88)
- Locally Optimal Private Sampling: Beyond the Global Minimax (NeurIPS 2025 · similarity 0.87)
- Certification for Differentially Private Prediction in Gradient-Based Training (ICML 2025 · similarity 0.87)
- Privacy Guarantees in Posterior Sampling under Contamination (NeurIPS 2026 · similarity 0.87)
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.
