Research map / Federated learning and privacy / Differential privacy
Tight Data Access Bounds for Private Top-$k$ Selection
Hao WU, Olga Ohrimenko, Anthony Wirth
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 both sorted access and random access (threshold algorithm over Gumbel/Laplace-noised scores) for private top-k selection | Private top-k selection (private threshold algorithm) · theoretical access cost analysis | expected access cost | Improved |
| Apply the exponential mechanism via the same sorted+random access algorithm (k=1, Gumbel noise) | Exponential mechanism · theoretical access cost analysis | expected access cost | Improved |
| Restrict the system to only random access or only sorted access (not both) | Private top-k selection lower bound · theoretical access cost lower bound | expected access cost: Omega(m) lower bound | Hurt |
| Pre-generating all m noise values replaced by online on-the-fly sampling of noisy tuples (ablation) | Private threshold algorithm (online sampling variant) · theoretical computation time analysis | expected time per access | No clear effect |
Closest papers by meaning
- Faster Differentially Private Top-$k$ Selection: A Joint Exponential Mechanism with Pruning (NeurIPS 2024 · similarity 0.90)
- Practical Differentially Private Top-k Selection with Pay-what-you-get Composition (NeurIPS 2019 · similarity 0.90)
- Oneshot Differentially Private Top-k Selection (ICML 2021 · similarity 0.90)
- A Joint Exponential Mechanism For Differentially Private Top-$k$ (ICML 2022 · similarity 0.89)
- Query-Efficient Locally Private Hypothesis Selection via the Scheffe Graph (NeurIPS 2025 · similarity 0.87)
- Sort, Partition, Randomize: Optimal Binary Hypothesis Testing under Local Differential Privacy (NeurIPS 2026 · similarity 0.86)
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.
