Research map / Federated learning and privacy / Differential privacy
Differentially Private Graph Learning via Sensitivity-Bounded Personalized PageRank
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 joint DP sensitivity-bounded PPR instead of edge-flipping randomized-response baseline | DPPushFlowCap (joint DP PPR) · POS and BlogCatalog PPR ranking | Recall@100 and NDCG | Helped |
| Use Laplace noise of scale 1/epsilon on PushFlow PPR values instead of sensitivity-bounded PPR | Non-private PushFlow with Laplace noise · PPR ranking | ranking accuracy | Hurt |
| Use joint DP PPR-based InstantEmbedding instead of baseline DP PPR followed by hashing | Joint DP InstantEmbedding · Node classification (Figure 2) | Micro F1 | Helped |
| Baseline joint DP embedding at small epsilon versus random embedding | Joint DP Baseline + InstantEmbedding · Node classification (Figure 2) | Micro F1 | No clear effect |
| Vary the sensitivity parameter sigma (ablation) | DPPushFlowCap · POS and BlogCatalog | utility | No clear effect |
Closest papers by meaning
- Differentially Private Graph Diffusion with Applications in Personalized PageRanks (NeurIPS 2024 · similarity 0.89)
- Going Deeper into Locally Differentially Private Graph Neural Networks (ICML 2025 · similarity 0.85)
- Poincaré Differential Privacy for Hierarchy-Aware Graph Embedding (AAAI 2024 · similarity 0.84)
- On Differential Privacy for Adaptively Solving Search Problems via Sketching (ICML 2025 · similarity 0.84)
- Accelerated Evolving Set Processes for Local PageRank Computation (NeurIPS 2025 · similarity 0.84)
- Certification for Differentially Private Prediction in Gradient-Based Training (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.