Research map / Federated learning and privacy / Differential privacy
Differentially Private Community Detection for Stochastic Block Models
Mohamed S Mohamed, Dung Nguyen, Anil Vullikanti, Ravi Tandon
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 MLE-based stability rather than graph perturbation with RR + SDP | Private community detection · Synthetic SBM, r=2 and r=3 | error probability | Improved |
| Use SDP-stability rather than RR + SDP | Private community detection · Political Blogosphere | error probability | Improved |
| Use SDP-stability rather than RR + SDP | Private community detection · Zachary's Karate Club | error probability | Mixed |
| Increase the separation between a and b | Private community detection mechanisms · Synthetic SBM | error probability | Improved |
| Increase ε | SDP-stability and RR + SDP · Synthetic SBM, r=2 | error probability | Improved |
| Use RR + SDP instead of RR-Spectral | Randomized-response recovery · Synthetic SBM, r=2 | error probability | Improved |
| Use finite ε with stability-based mechanisms rather than requiring ε to grow with n | Stability-based community recovery · Synthetic SBM | exact recovery | Improved |
| Use SDP-stability rather than RR + SDP | Private community detection · Synthetic SBM, r=3 | error probability | Improved |
| Use RR + SDP rather than RR-Spectral | Randomized-response recovery · Synthetic SBM, r=2 | computational complexity | Hurt |
| Use pure-DP sampling mechanisms rather than stability-based mechanisms | Private community recovery · SBM exact recovery | recovery threshold | Hurt |
Closest papers by meaning
- Differentially private exact recovery for stochastic block models (ICML 2024 · similarity 0.93)
- A Differentially Private Clustering Algorithm for Well-Clustered Graphs (ICLR 2024 · similarity 0.88)
- Differentially Private Gomory-Hu Trees (NeurIPS 2025 · similarity 0.88)
- Exact Community Recovery under Side Information: Optimality of Spectral Algorithms (ICLR 2025 · similarity 0.87)
- On the Price of Differential Privacy for Hierarchical Clustering (ICLR 2025 · similarity 0.87)
- Harnessing Multiple Correlated Networks for Exact Community Recovery (NeurIPS 2024 · 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.
