Research map / Federated learning and privacy / Differential privacy
Differentially Private Quantiles
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 |
|---|---|---|---|
| Estimate all requested quantiles jointly with JointExp instead of composing independent quantile estimates | Quantile estimation · Gaussian, uniform, Goodreads ratings, and Goodreads page counts | missed points: roughly 2-3 times smaller | Helped |
| Use JointExp rather than AppIndExp | Quantile estimation · All four datasets, m=1 | missed points | No clear effect |
| Use JointExp rather than CSmooth or AggTree | Quantile estimation · All four datasets, m=1 | missed points: roughly an order of magnitude better error | Helped |
| Use JointExp rather than the closest competitor | Quantile estimation · All four datasets, around m=30 | missed points | No clear effect |
| Run JointExp with the dynamic-programming sampler | JointExp · Gaussian data, m=30, n=1,000 | runtime: roughly 1 ms | Helped |
| Run JointExp with the dynamic-programming sampler | JointExp · Gaussian data, m=30, n=1 million | runtime: slightly under 1 minute | Mixed |
Closest papers by meaning
- Differentially Private Quantiles with Smaller Error (NeurIPS 2025 · similarity 0.93)
- Differentially Private Approximate Quantiles (ICML 2022 · similarity 0.93)
- Lightweight Protocols for Distributed Private Quantile Estimation (ICML 2025 · similarity 0.91)
- Differentially Private Space-Efficient Algorithms for Counting Distinct Elements in the Turnstile Model (ICML 2025 · similarity 0.86)
- Online Differentially Private Conformal Prediction for Uncertainty Quantification (ICML 2025 · similarity 0.86)
- Differentially Private Equivalence Testing for Continuous Distributions and Applications (NeurIPS 2024 · similarity 0.85)
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.