EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Evaluation of Open-Source Tools for Differential Privacy

Shiliang Zhang, Anton Hagermalm, Sanjin Slavnic, Elad Michael Schiller iD, Magnus Almgren iD

Sensors · 19.07.2023 · 23 (14) · 6509

DOI: 10.3390/s23146509 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 19.07.2023. Verfasst von Shiliang Zhang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Computer Science and Engineering, Chalmers University of Technology, SE-41296 Gothenburg, Sweden. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Area of Advance Transport and Chalmers AI Research Centre (CHAIR) at Chalmers Univ. Tech., VINNOVA, Area of Advance Transport and Chalmers AI Research Centre (CHAIR).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Differential privacy (DP) defines privacy protection by promising quantified indistinguishability between individuals who consent to share their privacy-sensitive information and those who do not. DP aims to deliver this promise by including well-crafted elements of random noise in the published data, and thus there is an inherent tradeoff between the degree of privacy protection and the ability to utilize the protected data. Currently, several open-source tools have been proposed for DP provision. To the best of our knowledge, there is no comprehensive study for comparing these open-source tools with respect to their ability to balance DP’s inherent tradeoff as well as the use of system resources. This work proposes an open-source evaluation framework for privacy protection solutions and offers evaluation for OpenDP Smartnoise, Google DP, PyTorch Opacus, Tensorflow Privacy, and Diffprivlib. In addition to studying their ability to balance the above tradeoff, we consider discrete and continuous attributes by quantifying their performance under different data sizes. Our results reveal several patterns that developers should have in mind when selecting tools under different application needs and criteria. This evaluation survey can be the basis for an improved selection of open-source DP tools and quicker adaptation of DP.

Schlagwörter

Stichwörter der Autor:innen

differential privacyevaluationopen-source tools

Autor:innen und Institutionen

  • Computer Science and Engineering, Chalmers University of Technology, SE-41296 Gothenburg, Sweden.
  • Department of Informatics, University of Oslo, NO-0373 Oslo, Norway.

ORCID-Profile: Elad Michael Schiller, Magnus Almgren

Förderung

  • Area of Advance Transport and Chalmers AI Research Centre (CHAIR) at Chalmers Univ. Tech.
  • VINNOVA
  • Area of Advance Transport and Chalmers AI Research Centre (CHAIR)
  • AstraZeneca AB
  • Swedish National Infrastructure for Computing (SNIC) at C3SE
  • Swedish Research Council

Zitieren

APA

Zhang, S., Hagermalm, A., Slavnic, S., Schiller, E. M., & Almgren, M. (2023). Evaluation of Open-Source Tools for Differential Privacy. Sensors, 23(14), 6509. https://doi.org/10.3390/s23146509

Vancouver

Zhang S, Hagermalm A, Slavnic S, Schiller EM, Almgren M. Evaluation of Open-Source Tools for Differential Privacy. Sensors. 2023;23(14):6509. doi:10.3390/s23146509

BibTeX
@article{Zhang2023,
  author = {Zhang, Shiliang and Hagermalm, Anton and Slavnic, Sanjin and Schiller, Elad Michael and Almgren, Magnus},
  title = {Evaluation of Open-Source Tools for Differential Privacy},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {14},
  pages = {6509},
  doi = {10.3390/s23146509}
}
Kostenlos starten →