Personalised Differential Privacy Summary of POPL ’ 15 paper “ Differential Privacy : Now It ’ s Getting Personal ”

By Hamid Ebadi, David Sands. 2015

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Advanced Probabilistic Couplings for Differential Privacy (2016, neutral): “…proposed for providing formal veriﬁcation of the differential privacy guarantee, using a wide variety of veriﬁcation approaches: dynamic checking [24, 36], relational program logic [2, 6] and relational reﬁnement type systems [9], linear (dependent) type systems [29, 42], product programs [7],…”
Proving Differential Privacy via Probabilistic Couplings (2016, neutral): “There are many language-based techniques for proving differential privacy for programs, including dynamic checking (McSherry 2009; Ebadi et al. 2015), the already mentioned relational program logic (Barthe et al.”
AutoPriv: Automating Differential Privacy Proofs (2016, methodology): “putation when the privacy budget is exhausted. Airavat (Roy et al. 2010) is a MapReduce-based system with a runtime monitor that enforces privacy policies controlled by data providers. Recent work by Ebadi et al. (2015) proposed Personalised Differential Privacy (PDP), where each individual has its ”
Optimizing error of high-dimensional statistical queries under differential privacy (2018, methodology): “One class of techniques answers the queries of interest on the database and then uses the Laplace Mechanism to add noise, calibrated to their sensitivity, or the maximum change in answers resulting from one change in the input database [11,13,21,40].”
Per-instance Differential Privacy and the Adaptivity of Posterior Sampling in Linear and Ridge regression (2017, methodology): “Data set private target probability metric parametrized by Pure-DP[10] supZ supz D∞(P‖Q) A only Approx-DP[8] supZ supz D ∞(P‖Q) A only (z/m)-CDP[12, 3] supZ supz DsubG(P‖Q) A only Rényi-DP[20] supZ supz Dα(P‖Q) A only Personal-DP[13, 17] supZ fixed z D ∞(P‖Q) A and z TV-privacy[2] supZ supz ‖P −Q‖TV”