Research map / Federated learning and privacy / Differential privacy
Collecting Telemetry Data Privately
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 1BitRRPM instead of Laplace | Mean estimation · Real-world app-usage data | mean absolute error | Helped |
| Add output perturbation with γ=1/10 | 1BitRRPM · Real-world app-usage data | mean absolute error | No clear effect |
| Add output perturbation with γ=1/3 | 1BitRRPM · Real-world app-usage data | mean absolute error | Hurt |
| Use 32-bit dBitFlipPM (d=k) | Histogram estimation · Real-world app-usage data, 32 buckets | histogram error | Helped |
| Use d=4 dBitFlipPM instead of KFlip | Histogram estimation · Real-world app-usage data, ε=0.1–0.5 | histogram error | No clear effect |
| Use d=1, 2, or 4 dBitFlipPM instead of BinFlip+ | Histogram estimation · Real-world app-usage data, ε=5–10 | histogram error | Helped |
| Add α-point randomized rounding before memoization (ablation) | 1BitMean with permanent memoization · Mean estimation, single collection round | mean estimation error | No clear effect |
| Add output perturbation (ablation) | 1BitMean with memoization · Mean estimation, single collection round | mean estimation error | Hurt |
| Use memoization with α-point rounding | Repeated mean collection · Synthetic datasets | mean estimation error | No clear effect |
| Use memoization with α-point rounding | Repeated histogram collection · Synthetic datasets | histogram error | No clear effect |
Closest papers by meaning
- Monitoring Violations of Differential Privacy over Time (NeurIPS 2026 · similarity 0.84)
- Lightweight Protocols for Distributed Private Quantile Estimation (ICML 2025 · similarity 0.83)
- Local Pan-privacy for Federated Analytics (ICML 2025 · similarity 0.83)
- Private and Personalized Frequency Estimation in a Federated Setting (NeurIPS 2024 · similarity 0.83)
- A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy (AAAI 2024 · similarity 0.83)
- Optimal Survey Design for Private Mean Estimation (ICML 2025 · similarity 0.83)
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.