Privacy Auditing with One (1) Training Run
Authors:
- Thomas SteinkeTSThomas Steinke
- Milad NasrMNMilad Nasr
- Matthew JagielskiMJMatthew Jagielski
Preprint
Abstract
We propose a scheme for auditing differentially private machine learning systems with a single training run. This exploits the parallelism of being able to add or remove multiple training examples independently. We analyze this using the connection between differential privacy and statistical generalization, which avoids the cost of group privacy. Our auditing scheme requires minimal assumptions about the algorithm and can be applied in the black-box or white-box setting.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Chapter 3 Composition of Differential Privacy & Privacy Amplification by Subsampling2025
- Wide Residual Networks2016
- Private Empirical Risk Minimization: Efficient Algorithms and Tight Error Bounds2014
- The Algorithmic Foundations of Differential PrivacyFoundations and Trends® in Theoretical Computer Science, 2014
- On the 'Semantics' of Differential Privacy: A Bayesian FormulationJournal of Privacy and Confidentiality, 2014
- Genomic privacy and limits of individual detection in a poolNature Genetics, 2009
- Resolving Individuals Contributing Trace Amounts of DNA to Highly Complex Mixtures Using High-Density SNP Genotyping MicroarraysPLOS Genetics, 2008
- Optimal probabilistic fingerprint codesJournal of the ACM, 2008
- Collusion-secure fingerprinting for digital dataIEEE Transactions on Information Theory, 1998