Differential private relevance learning
Brinkrolf J, Berger K, Hammer B (2018)
In: Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018). Verleysen M (Ed); 555-560.
Konferenzbeitrag
| Veröffentlicht | Englisch
Download
Es wurden keine Dateien hochgeladen. Nur Publikationsnachweis!
Autor*in
Herausgeber*in
Verleysen, Michel
Einrichtung
Abstract / Bemerkung
Digital information is collected daily in growing volumes.
Mutual benefits drive the demand for the exchange and publication of
data among parties. However, it is often unclear how to handle these data
properly in the case that the data contains sensitive information. Differ-
ential privacy has become a powerful principle for privacy-preserving data
analysis tasks in the last few years, since it entails a formal privacy guar-
antee for such settings. This is obtained by a separation of the utility of
the database and the risk of an individual to lose his/her privacy. In this
contribution, we introduce the Laplace mechanism and a stochastic gradient
descent methodology which guarantee differential privacy [1]. Then,
we show how these paradigms can be incorporated into two popular ma-
chine learning algorithm, namely GLVQ and GMLVQ. We demonstrate
the results of privacy-preserving LVQ based on three benchmarks.
Erscheinungsjahr
2018
Titel des Konferenzbandes
Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018)
Seite(n)
555-560
Konferenz
26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)
Konferenzort
Bruges (Belgium)
Konferenzdatum
2018-04-25 – 2018-04-27
ISBN
978-287587047-6
Page URI
https://pub.uni-bielefeld.de/record/2918254
Zitieren
Brinkrolf J, Berger K, Hammer B. Differential private relevance learning. In: Verleysen M, ed. Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018). 2018: 555-560.
Brinkrolf, J., Berger, K., & Hammer, B. (2018). Differential private relevance learning. In M. Verleysen (Ed.), Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018) (pp. 555-560).
Brinkrolf, Johannes, Berger, Kolja, and Hammer, Barbara. 2018. “Differential private relevance learning”. In Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018), ed. Michel Verleysen, 555-560.
Brinkrolf, J., Berger, K., and Hammer, B. (2018). “Differential private relevance learning” in Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018), Verleysen, M. ed. 555-560.
Brinkrolf, J., Berger, K., & Hammer, B., 2018. Differential private relevance learning. In M. Verleysen, ed. Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018). pp. 555-560.
J. Brinkrolf, K. Berger, and B. Hammer, “Differential private relevance learning”, Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018), M. Verleysen, ed., 2018, pp.555-560.
Brinkrolf, J., Berger, K., Hammer, B.: Differential private relevance learning. In: Verleysen, M. (ed.) Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018). p. 555-560. (2018).
Brinkrolf, Johannes, Berger, Kolja, and Hammer, Barbara. “Differential private relevance learning”. Proceedings of the 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2018). Ed. Michel Verleysen. 2018. 555-560.
Link(s) zu Volltext(en)
Access Level
Open Access
Google Scholar
ISBN Suche