EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Differentially Private Confidence Intervals for Empirical Risk Minimization

Yue Wang, Daniel Kifer, Jaewoo Lee

Journal of Privacy and Confidentiality · 2019

DOI: 10.29012/jpc.660 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Privacy and Confidentiality, 2019. Verfasst von Yue Wang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The process of data mining with differential privacy produces results that are affected by two types of noise: sampling noise due to data collection and privacy noise that is designed to prevent the reconstruction of sensitive information. In this paper, we consider the problem of designing confidence intervals for the parameters of a variety of differentially private machine learning models. The algorithms can provide confidence intervals that satisfy differential privacy (as well as the more recently proposed concentrated differential privacy) and can be used with existing differentially private mechanisms that train models using objective perturbation and output perturbation.

Zitieren

APA

Wang, Y., Kifer, D., & Lee, J. (2019). Differentially Private Confidence Intervals for Empirical Risk Minimization. Journal of Privacy and Confidentiality. https://doi.org/10.29012/jpc.660

Vancouver

Wang Y, Kifer D, Lee J. Differentially Private Confidence Intervals for Empirical Risk Minimization. Journal of Privacy and Confidentiality. 2019. doi:10.29012/jpc.660

BibTeX
@article{Wang2019,
  author = {Wang, Yue and Kifer, Daniel and Lee, Jaewoo},
  title = {Differentially Private Confidence Intervals for Empirical Risk Minimization},
  journal = {Journal of Privacy and Confidentiality},
  year = {2019},
  doi = {10.29012/jpc.660}
}
Kostenlos starten →