Understanding Disclosure Risk in Differential Privacy with Applications to Noise Calibration and Auditing
Summary: Introduces reconstruction advantage, a unified disclosure-risk metric for DP that subsumes membership, attribute inference, and reconstruction. Derives tight noise-to-risk bounds and optimal attacks, enabling principled noise calibration and systematic DP auditing beyond ReRo. (summarized by gpt-5.4-mini on May 27 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Patricia Guerra-Balboa (Karlsruhe Institute of Technology)
- 2. Annika Sauer (Karlsruhe Institute of Technology)
- 3. Héber H. Arcolezi (INRIA; École de technologie supérieure, Montreal)
- 4. Thorsten Strufe (Karlsruhe Institute of Technology)
BibTeX Citation
@article{guerrabalboa_vldb26,
title = {{Understanding Disclosure Risk in Differential Privacy with Applications to Noise Calibration and Auditing}},
author = {Guerra-Balboa, Patricia and Sauer, Annika and Arcolezi, Héber H. and Strufe, Thorsten},
journal = {PVLDB},
series = {{VLDB} '26},
volume = {19},
number = {7},
pages = {1558--1571},
doi = {10.14778/3801059.3801069},
url = {https://doi.org/10.14778/3801059.3801069},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 7 of 7 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 62 | Privacy Integrated Queries: An Extensible Platform for Privacy-Preserving Data Analysis | 2009 | SIGMOD | 0.00038970535 |
| 5,327 | Real-World Trajectory Sharing with Local Differential Privacy | 2021 | VLDB | 6.2639736e-05 |
| 7,841 | Quantifying identifiability to choose and audit epsilon in differentially private deep learning | 2021 | VLDB | 5.5335409e-05 |
| 8,514 | Synthetic Tabular Data: Methods, Attacks and Defenses | 2025 | VLDB | 5.4119882e-05 |
| 8,515 | Measuring Re-identification Risk | 2023 | SIGMOD | 5.4119882e-05 |
| 8,548 | Privacy Preserving Serial Data Publishing By Role Composition | 2008 | VLDB | 5.4119882e-05 |
| 11,428 | On the Risks of Collecting Multidimensional Data Under Local Differential Privacy | 2023 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 7,456 | Privacy Amplification by Sampling under User-level Differential Privacy | 2024 | SIGMOD |
| 2 | 836 | Recursive Mechanism: Towards Node Differential Privacy and Unrestricted Joins | 2013 | SIGMOD |
| 3 | 3,296 | Towards Robustness in Query Auditing | 2006 | VLDB |
| 4 | 775 | Differentially Private Data Cubes: Optimizing Noise Sources and Consistency | 2011 | SIGMOD |
| 5 | 4,531 | Distributed Privacy Preserving Information Sharing | 2005 | VLDB |
| 6 | 10,335 | A General Framework for Per-record Differential Privacy | 2026 | SIGMOD |
| 7 | 7,841 | Quantifying identifiability to choose and audit epsilon in differentially private deep learning | 2021 | VLDB |
| 8 | 6,441 | Differential Privacy in Data Publication and Analysis | 2012 | SIGMOD |
| 9 | 10,451 | Enhancing Local Differential Privacy Accuracy by Exploiting Inherent Uncertainty | 2026 | SIGMOD |
| 10 | 10,786 | Computing Inconsistency Measures Under Differential Privacy | 2025 | SIGMOD |