Research map / Federated learning and privacy / Differential privacy
Geographic Differential Privacy for Mobile Crowd Coverage Maximization
Leye Wang, Gehua Qin, Dingqi Yang, Xiao Han, Xiaojuan Ma
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 the optimized geographic differential privacy policy instead of Laplace obfuscation | Mobile crowd coverage selection · FS, single-location and multi-location coverage | coverage probability: up to 5% improvement | Improved |
| Use the optimized geographic differential privacy policy instead of Laplace obfuscation | Mobile crowd coverage selection · CMCC and D4D, multi-location coverage | coverage probability | Improved |
| Increase the number of target locations covered (ablation) | Proposed privacy mechanism · FS, CMCC, and D4D | coverage probability relative to No-Obfuscation | Improved |
| Increase privacy budget ε, thereby weakening privacy protection (ablation) | Proposed privacy mechanism · FS, dense single target | coverage probability | Improved |
| Use Poisson-based instead of frequency-based mobility profiling (ablation) | Mobile crowd coverage selection · FS, CMCC, and D4D | AUC | Improved |
| Increase the frequent-location threshold δ (ablation) | Proposed privacy mechanism · FS, single-location coverage | coverage probability | Improved |
| Increase the frequent-location threshold δ (ablation) | Proposed privacy mechanism · FS, candidate user availability | number of users eligible to upload frequent locations | Hurt |
| Increase the number of user groups k for iterative estimation of π (ablation) | Proposed user-selection algorithm · FS, estimating π | KL divergence | Improved |
| Increase the number of user groups k for iterative estimation of π (ablation) | Proposed user-selection algorithm · FS, algorithm runtime | runtime | Hurt |
| Apply geographic differential privacy rather than upload real frequent locations | Proposed method · FS, dense single target | coverage probability: 32.9% -> 21.7% | Hurt |
| Apply geographic differential privacy rather than upload real frequent locations | Proposed method · FS, less-dense single target | coverage probability: 30.5% -> 14.5% | Hurt |
Closest papers by meaning
- A Generalized Shuffle Framework for Privacy Amplification: Strengthening Privacy Guarantees and Enhancing Utility (AAAI 2024 · similarity 0.84)
- A Huber Loss Minimization Approach to Mean Estimation under User-level Differential Privacy (NeurIPS 2024 · similarity 0.82)
- A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy (AAAI 2024 · similarity 0.82)
- Privacy amplification by random allocation (NeurIPS 2025 · similarity 0.81)
- Optimal Survey Design for Private Mean Estimation (ICML 2025 · similarity 0.81)
- Locally Optimal Private Sampling: Beyond the Global Minimax (NeurIPS 2025 · similarity 0.80)
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.
