Random Sampling Plus Fake Data: Multidimensional Frequency Estimates With Local Differential Privacy - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Random Sampling Plus Fake Data: Multidimensional Frequency Estimates With Local Differential Privacy

Résumé

With local differential privacy (LDP), users can privatize their data and thus guarantee privacy properties before transmitting it to the server (a.k.a. the aggregator). One primary objective of LDP is frequency (or histogram) estimation, in which the aggregator estimates the number of users for each possible value. In practice, when a study with rich content on a population is desired, the interest is in the multiple attributes of the population, that is to say, in multidimensional data ($d \geq 2$). However, contrary to the problem of frequency estimation of a single attribute (the majority of the works), the multidimensional aspect imposes to pay particular attention to the privacy budget. This one can indeed grow extremely quickly due to the composition theorem. To the authors' knowledge, two solutions seem to stand out for this task: 1) splitting the privacy budget for each attribute, i.e., send each value with $\frac{\epsilon}{d}$-LDP (\textit{Spl}), and 2) random sampling a single attribute and spend all the privacy budget to send it with $\epsilon$-LDP (\textit{Smp}). Although \textit{Smp} adds additional sampling error, it has proven to provide higher data utility than the former \textit{Spl} solution. However, we argue that aggregators (who are also seen as attackers) are aware of the sampled attribute and its LDP value, which is protected by a "less strict" $\epsilon$ probability bound (rather than $\frac{\epsilon}{d}$). This way, we propose a solution named \underline{R}andom \underline{S}ampling plus \underline{F}ake \underline{D}ata (RS+FD), which allows creating \textit{uncertainty} over the sampled attribute by generating fake data for each non-sampled attribute; RS+FD further benefits from amplification by sampling. We theoretically and experimentally validate our proposed solution on both synthetic and real-world datasets to show that RS+FD achieves nearly the same or better utility than the state-of-the-art \textit{Smp} solution.
Fichier principal
Vignette du fichier
4c4c028e-c585-407a-87b6-613a6a2f2609-author.pdf (961.35 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03551918 , version 1 (02-02-2022)

Identifiants

  • HAL Id : hal-03551918 , version 1

Citer

Jean-François Couchot, Héber Hwang Arcolezi, Bechara Al Bouna, Xiaokui Xiao. Random Sampling Plus Fake Data: Multidimensional Frequency Estimates With Local Differential Privacy. International Conference on Information and Knowledge Management, Nov 2021, online, Australia. ⟨hal-03551918⟩
92 Consultations
54 Téléchargements

Partager

Gmail Facebook X LinkedIn More