Research map / Federated learning and privacy / Differential privacy
Estimating the True Distribution of Data Collected with Randomized Response
Carlos Antonio Pinzón, Ehab ElSalamouny, Lucas Massot, Alexis Miller, Héber Hwang Arcolezi, Catuscia Palamidessi
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 |
|---|---|---|---|
| stronger privacy (epsilon 1.0 -> 0.5) slows IBU convergence (ablation) | IBU · RR, K=500, n=10^4 | iterations to match MLE*: 7000 -> 25000 | Hurt |
| larger sample size slows IBU convergence (ablation) | IBU · RR, K=500, epsilon=1.0 | iterations to match MLE*: 7000 -> 20000 | Hurt |
| use MLE* instead of InvP/InvN under a near-uniform distribution | RR estimators · Zipf s=0.01, K=10000, epsilon=4 | MSE | Mixed |
| use MLE* instead of InvN under a highly concentrated distribution | RR estimators · Zipf s=2.5, K=10000, epsilon=4 | MSE | Mixed |
| use MLE* instead of InvP/InvN | RR estimators · Zipf s in {0.01, 1.3, 2.5}, K=10000 | negative log-likelihood | Improved |
| use MLE* instead of InvP/InvN on real data | RR estimators · Kosarak, ACSIncome (PUMA, PINCP) | MSE | No clear effect |
| InvP vs InvN under moderately skewed data | RR post-processing · ACSIncome | MSE | Mixed |
Closest papers by meaning
- Differentially Private Analysis for Binary Response Models: Optimality, Estimation, and Inference (ICML 2025 · similarity 0.83)
- Analysis of Differentially Private Synthetic Data: A Measurement Error Approach (AAAI 2024 · similarity 0.80)
- Debiasing Synthetic Data Generated by Deep Generative Models (NeurIPS 2024 · similarity 0.80)
- A Principled Approach to Randomized Selection under Uncertainty: Applications to Peer Review and Grant Funding (NeurIPS 2025 · similarity 0.79)
- Improving the Variance of Differentially Private Randomized Experiments through Clustering (ICML 2025 · similarity 0.79)
- A Provably Accurate Randomized Sampling Algorithm for Logistic Regression (AAAI 2024 · similarity 0.78)
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.
