Quantifying identifiability to choose and audit epsilon in differentially private deep learning
Summary: Transforms (epsilon, delta) into a bound on posterior identifiability of training records under composition. Auditable DP adversary estimates identifiability; links to membership inference, enabling empirical (epsilon, delta) for parameter choice. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Daniel Bernau
- 2. Günther Eibl
- 3. Philip W. Grassal
- 4. Hannah Keller
- 5. Florian Kerschbaum
Incoming Citations (Sorted by Pagerank)
Showing 2 of 2 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,262 | Understanding Disclosure Risk in Differential Privacy with Applications to Noise Calibration and Auditing | 2026 | VLDB | 4.1945683e-05 |
| 10,499 | Privacy and Accuracy-Aware AI/ML Model Deduplication | 2025 | SIGMOD | 4.1945683e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 0 of 0 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Semantically Similar Papers
| Overall Rank | Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 5,687 | Differential Privacy in the Wild: A Tutorial on Current Practices & Open Challenges | 2017 | SIGMOD | 5.3706593e-05 |
| 8,438 | Data Publishing against Realistic Adversaries | 2009 | VLDB | 4.5131088e-05 |
| 10,262 | Understanding Disclosure Risk in Differential Privacy with Applications to Noise Calibration and Auditing | 2026 | VLDB | 4.1945683e-05 |
| 10,499 | Privacy and Accuracy-Aware AI/ML Model Deduplication | 2025 | SIGMOD | 4.1945683e-05 |
| 3,097 | Publishing Set-Valued Data via Differential Privacy | 2011 | VLDB | 7.5647028e-05 |
| 6,486 | Differential Privacy in Data Publication and Analysis | 2012 | SIGMOD | 5.0445043e-05 |
| 10,041 | A General Framework for Per-record Differential Privacy | 2026 | SIGMOD | 4.1945683e-05 |
| 10,162 | Enhancing Local Differential Privacy Accuracy by Exploiting Inherent Uncertainty | 2026 | SIGMOD | 4.1945683e-05 |
| 3,304 | Plausible Deniability for Privacy-Preserving Data Synthesis | 2017 | VLDB | 7.2467347e-05 |
| 3,172 | Bayesian Differential Privacy on Correlated Data | 2015 | SIGMOD | 7.4411955e-05 |