Research map / Federated learning and privacy / Differential privacy
Privately detecting changes in unknown distributions
Rachel Cummings, Sara Krehbiel, Yuliia Lut, Wanrong Zhang
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 |
|---|---|---|---|
| Increase the distributional shift from μ1=1 to μ1=5 | PNCPD · Synthetic Gaussian offline change-point detection | change-point estimation accuracy | Improved |
| Increase ε | PNCPD · Synthetic Gaussian offline change-point detection | change-point estimation accuracy | Improved |
| Place the change-point at the center of the window rather than nearer an edge | PNCPD · Synthetic Gaussian offline change-point detection | change-point estimation accuracy | Hurt |
| Use ε=1 rather than ε=5 or ε=10 | OnlinePNCPD · Synthetic Gaussian online change-point detection | inaccurate estimation and false-alarm rate | Hurt |
| Condition evaluation on halting in a window that contains the true change-point | OnlinePNCPD · Synthetic Gaussian online change-point detection | inaccurate estimation rate | Improved |
| Set the online threshold to T=0.8 despite the theoretical interval being empty | OnlinePNCPD · Synthetic Gaussian online change-point detection, ε=1 and ε=5 | false-alarm rates | Improved |
| Transform consecutive observations into pairwise differences before applying PNCPD | PNCPD for drift change detection · Synthetic linear-drift change detection | change-point estimation accuracy | Improved |
Closest papers by meaning
- Differentially Private Equivalence Testing for Continuous Distributions and Applications (NeurIPS 2024 · similarity 0.85)
- Monitoring Violations of Differential Privacy over Time (NeurIPS 2026 · similarity 0.84)
- Making Old Things New: A Unified Algorithm for Differentially Private Clustering (ICML 2024 · similarity 0.84)
- A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy (AAAI 2024 · similarity 0.83)
- Reducing sequential change detection to sequential estimation (ICML 2024 · similarity 0.83)
- Privacy Guarantees in Posterior Sampling under Contamination (NeurIPS 2026 · 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.
