Mechanism Design via Differential Privacy

By Frank McSherry, Kunal Talwar. 2007

360 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Science Applications, Management Science and Operations Research

How researchers see this paper

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

In Artificial Intelligence

In Computer Science Applications

In Management Science and Operations Research

Citations in the corpus

Proving Differential Privacy via Probabilistic Couplings (2016, positive): “In this section, we provide a formal proof of the Exponential mechanism of McSherry and Talwar [27].”
Differentially Private Empirical Risk Minimization (2009, positive): “In the notation of McSherry and Talwar (2007), the zmin = OPT , the base measure µ is uniform on [m], and St = {i : zi zmin + t}.”
Knowledge-based Systems (2009, neutral): “8 Recall from Section 3.2 that this case considers 34 European countries, two values (religiosity (rl) and permissiveness (pr)), and two actions (divorce (dv) and parenthood by homosexual couples (ho)).”
Adaptive Learning with Robust Generalization Guarantees (2016, positive): “The key results we use here are the upper bounds for private learning proven by [KLN + 11] using the exponential mechanism of [MT07], and the lower bounds for private proper threshold learning due to [BNSV15].”
GUPT: privacy preserving data analysis made easy (2012, neutral): “PINQ (Privacy INtegrated Queries) proposed programming constructs which enable application developers to write diﬀerentially private programs using basic functional building blocks of diﬀerential privacy ( e.g. , exponential mechanism [17], noisy counts [5] etc.).”
Plausible Deniability for Privacy-Preserving Data Synthesis (2017, neutral): “However, known mechanisms, such as the Laplacian mechanism [15] or the exponential mechanism [37], that achieve diﬀerential privacy through randomization, have practical limitations.”
The Limits of Two-Party Differential Privacy (2010, neutral): “Given a deterministic protocol for computing the sensitivity1 function f(x, y) we construct an -differentially-private protocol by sampling messages of the new protocol using the exponential mechanism [29].”
Differentially Private Online Learning (2011, neutral): “However, just for completeness, we note that by using techni ques similar to our Private OCP framework and using exponential mechanism(see [30]), one can analyze this algorithm as well to guarant ee differential privacy along with Õ( √ T ) regret.”
A Program Logic for Union Bounds (2016, neutral): “Report-noisy-max is a variant of the exponential mechanism [32], which provides the standard way to achieve diﬀerential privacy for computations whose outputs lie in a ﬁnite (perhaps non-numeric) set R .”
DJoin: Differentially Private Join Queries over Distributed Databases (2012, neutral): “, via the exponential mechanism [24], but we have not yet found a way to express them in terms of set intersections.”
Optimal data-independent noise for differential privacy (2013, neutral): “Other methods for attaining ε -diﬀerential privacy not based on directly adding noise to the real query response are, for instance, the exponential mechanism [14], and the sample and aggregate framework [16].”