Synthesizing differentially private datasets using random mixing

Cited 3 time in webofscience Cited 2 time in scopus
  • Hit : 159
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorLee, Kangwookko
dc.contributor.authorKim, Hoonko
dc.contributor.authorLee, Kyoungminko
dc.contributor.authorSuh, Changhoko
dc.contributor.authorRamchandran, Kannanko
dc.date.accessioned2019-12-13T10:27:10Z-
dc.date.available2019-12-13T10:27:10Z-
dc.date.created2019-11-26-
dc.date.created2019-11-26-
dc.date.created2019-11-26-
dc.date.created2019-11-26-
dc.date.issued2019-07-05-
dc.identifier.citationProceedings of the IEEE International Symposium on Information Theory(ISIT), pp.542 - 546-
dc.identifier.urihttp://hdl.handle.net/10203/269354-
dc.description.abstractThe goal of differentially private data publishing is to release a modified dataset so that its privacy can be ensured while allowing for efficient learning. We propose a new data publishing algorithm in which a released dataset is formed by mixing l randomly chosen data points and then perturbing them with an additive noise. Our privacy analysis shows that as l increases, noise with smaller variance is sufficient to achieve a target privacy level. In order to quantify the usefulness of our algorithm, we adopt the accuracy of a predictive model trained with our synthetic dataset, which we call the utility of the dataset. By characterizing the utility of our dataset as a function of l, we show that one can learn both linear and nonlinear predictive models so that they yield reasonably good prediction accuracies. Particularly, we show that there exists a sweet spot on l that maximizes the prediction accuracy given a required privacy level, or vice versa. We also demonstrate that given a target privacy level, our datasets can achieve higher utility than other datasets generated with the existing data publishing algorithms.-
dc.languageEnglish-
dc.publisherProceedings of the IEEE International Symposium on Information Theory-
dc.titleSynthesizing differentially private datasets using random mixing-
dc.typeConference-
dc.identifier.wosid000489100300110-
dc.identifier.scopusid2-s2.0-85073169963-
dc.type.rimsCONF-
dc.citation.beginningpage542-
dc.citation.endingpage546-
dc.citation.publicationnameProceedings of the IEEE International Symposium on Information Theory(ISIT)-
dc.identifier.conferencecountryFR-
dc.identifier.conferencelocationParis, France-
dc.identifier.doi10.1109/ISIT.2019.8849381-
dc.contributor.localauthorSuh, Changho-
dc.contributor.nonIdAuthorLee, Kangwook-
dc.contributor.nonIdAuthorKim, Hoon-
dc.contributor.nonIdAuthorLee, Kyoungmin-
dc.contributor.nonIdAuthorRamchandran, Kannan-
Appears in Collection
EE-Conference Papers(학술회의논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 3 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0