Max-Information, Differential Privacy, and Post-selection Hypothesis Testing

@article{Rogers2016MaxInformationDP,
  title={Max-Information, Differential Privacy, and Post-selection Hypothesis Testing},
  author={Ryan M. Rogers and Aaron Roth and Adam D. Smith and Om Thakkar},
  journal={2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS)},
  year={2016},
  pages={487-494}
}
In this paper, we initiate a principled study of how the generalization properties of approximate differential privacy can be used to perform adaptive hypothesis testing, while giving statistically valid p-value corrections. We do this by observing that the guarantees of algorithms with bounded approximate max-information are sufficient to correct the p-values of adaptively chosen hypotheses, and then by proving that algorithms that satisfy (∈,δ)-differential privacy have bounded approximate… CONTINUE READING

Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 26 CITATIONS

Leveraging Privacy in Data Analysis

VIEW 8 EXCERPTS
CITES BACKGROUND & METHODS

Information Theory of Data Privacy

VIEW 6 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

On relaxed security notions for secret sharing

  • IACR Cryptology ePrint Archive
  • 2019
VIEW 2 EXCERPTS
CITES BACKGROUND & METHODS

References

Publications referenced by this paper.