Privacy Amplification of Iterative Algorithms via Contraction Coefficients

@article{Asoodeh2020PrivacyAO,
  title={Privacy Amplification of Iterative Algorithms via Contraction Coefficients},
  author={S. Asoodeh and M. D{\'i}az and F. Calmon},
  journal={2020 IEEE International Symposium on Information Theory (ISIT)},
  year={2020},
  pages={896-901}
}
  • S. Asoodeh, M. Díaz, F. Calmon
  • Published 2020
  • Mathematics, Computer Science
  • 2020 IEEE International Symposium on Information Theory (ISIT)
We investigate the framework of privacy amplification by iteration, recently proposed by Feldman et al., from an information-theoretic lens. We demonstrate that differential privacy guarantees of iterative mappings can be determined by a direct application of contraction coefficients derived from strong data processing inequalities for f-divergences. In particular, by generalizing the Dobrushin’s contraction coefficient for total variation distance to an f-divergence known as Eγ-divergence, we… Expand
7 Citations
Privacy Analysis of Online Learning Algorithms via Contraction Coefficients
  • 1
  • PDF
A Compression Perspective on Secrecy Measures
Secrecy by Design With Applications to Privacy and Compression
  • 3
  • PDF
the Connection between Cryptography and Differential Privacy: a Survey
  • PDF

References

SHOWING 1-10 OF 39 REFERENCES
Privacy Amplification by Iteration
  • 56
  • Highly Influential
  • PDF
Concentrated Differential Privacy: Simplifications, Extensions, and Lower Bounds
  • 281
  • PDF
Privacy Amplification by Subsampling: Tight Analyses via Couplings and Divergences
  • 78
  • PDF
Privacy Amplification by Mixing and Diffusion Mechanisms
  • 12
  • Highly Influential
  • PDF
Rényi Differential Privacy
  • Ilya Mironov
  • Computer Science
  • 2017 IEEE 30th Computer Security Foundations Symposium (CSF)
  • 2017
  • 343
  • PDF
The Composition Theorem for Differential Privacy
  • 251
  • PDF
Differentially Private Empirical Risk Minimization
  • 836
  • PDF
...
1
2
3
4
...