Corpus ID: 203593788

Optimal Differential Privacy Composition for Exponential Mechanisms and the Cost of Adaptivity

@article{Dong2019OptimalDP,
  title={Optimal Differential Privacy Composition for Exponential Mechanisms and the Cost of Adaptivity},
  author={Jinshuo Dong and D. Durfee and R. Rogers},
  journal={ArXiv},
  year={2019},
  volume={abs/1909.13830}
}
  • Jinshuo Dong, D. Durfee, R. Rogers
  • Published 2019
  • Computer Science
  • ArXiv
  • Composition is one of the most important properties of differential privacy (DP), as it allows algorithm designers to build complex private algorithms from DP primitives. We consider precise composition bounds of the overall privacy loss for exponential mechanisms, one of the fundamental classes of mechanisms in DP. We give explicit formulations of the optimal privacy loss for both the adaptive and non-adaptive settings. For the non-adaptive setting in which each mechanism has the same privacy… CONTINUE READING
    3 Citations
    Tight Approximate Differential Privacy for Discrete-Valued Mechanisms Using FFT
    • 1
    Deep Learning with Gaussian Differential Privacy
    • 26
    • PDF
    LinkedIn's Audience Engagements API: A Privacy Preserving Data Analytics System at Scale
    • 7
    • PDF

    References

    SHOWING 1-10 OF 17 REFERENCES
    The Complexity of Computing the Optimal Composition of Differential Privacy
    • 8
    • Highly Influential
    Mechanism Design via Differential Privacy
    • F. McSherry, Kunal Talwar
    • Computer Science
    • 48th Annual IEEE Symposium on Foundations of Computer Science (FOCS'07)
    • 2007
    • 1,520
    • Highly Influential
    • PDF
    Gaussian Differential Privacy
    • 30
    • PDF
    The Composition Theorem for Differential Privacy
    • 227
    • PDF
    Is Interaction Necessary for Distributed Private Learning?
    • 90
    • PDF
    The Role of Interactivity in Local Differential Privacy
    • 28
    • PDF
    Practical Differentially Private Top-k Selection with Pay-what-you-get Composition
    • 13
    • PDF
    Privacy Odometers and Filters: Pay-as-you-Go Composition
    • 36
    • PDF
    Lower Bounds for Locally Private Estimation via Communication Complexity
    • 35
    • PDF
    Concentrated Differential Privacy: Simplifications, Extensions, and Lower Bounds
    • 250
    • PDF