Unconditional differentially private mechanisms for linear queries

By Aditya Bhaskara, Daniel Dadush, Ravishankar Krishnaswamy, Kunal Talwar. 2012

57 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Science Applications

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

Citations in the corpus

The structure of optimal private tests for simple hypotheses (2018, background): “There are also several rich lines of work attempting to give tight instance-specific characterizations of the sample complexities of various differentially private computations, most notably linear query release [11, 43, 48, 57, 58] and PAC and agnostic learning [10, 39, 46].”
Lower Bounds for Differential Privacy from Gaussian Width (2016, methodology): “This object has been important in applying geometric techniques to diﬀerential privacy [9, 2, 16, 15].”
Protecting Locations with Differential Privacy under Temporal Correlations (2014, methodology): “Because we proved (in Lemma 4.1) that a query matrix is equal to a polytope (polygon if in two-dimensional space), state-of-art query-answering method, K -norm based mechanism [9, 26, 42], can be used for our problem.”
On the relation between identifiability, differential privacy, and mutual-information privacy (2014, background): “mness is added to the database to generate a synthetic database. For the interactive model, a signiﬁcant body of work has also been devoted to privacy–usefulness tradeoff for speciﬁc types of queries [10, 2, 17, 5, 18], the convergence rates of privacy-preserving statistical estimation have been stu”
Optimizing error of high-dimensional statistical queries under differential privacy (2018, background): “There has been a plethora of work on differentially private techniques for answering sets of queries including work establishing theoretical lower bounds [4,17,31] and practical algorithms [2, 3, 8, 10,19,24–26,28,33–35,35,41–50].”
The Complexity of Differential Privacy (2017, methodology): “If we work with `2 rather than `∞ error, then tighter characterizations of the rate of error decay are known (up to factors polylog(|Q|, |X|)), by applying more sophisticated geometric methods to the convex body K [59, 11, 85].”
Iterative Constructions and Private Data Release (2011, unknown): “Subsequent improvements were given by [3].”