Detecting violations of differential privacy
The widespread acceptance of differential privacy has led to the publication of many
sophisticated algorithms for protecting privacy. However, due to the subtle nature of this …
sophisticated algorithms for protecting privacy. However, due to the subtle nature of this …
Reproducibility in learning
R Impagliazzo, R Lei, T Pitassi, J Sorrell - Proceedings of the 54th annual …, 2022 - dl.acm.org
We introduce the notion of a reproducible algorithm in the context of learning. A reproducible
learning algorithm is resilient to variations in its samples—with high probability, it returns the …
learning algorithm is resilient to variations in its samples—with high probability, it returns the …
Dp-sniper: Black-box discovery of differential privacy violations using classifiers
We present DP-Sniper, a practical black-box method that automatically finds violations of
differential privacy. DP-Sniper is based on two key ideas:(i) training a classifier to predict if …
differential privacy. DP-Sniper is based on two key ideas:(i) training a classifier to predict if …
Group and attack: Auditing differential privacy
(ε, δ) differential privacy has seen increased adoption recently, especially in private machine
learning applications. While this privacy definition allows provably limiting the amount of …
learning applications. While this privacy definition allows provably limiting the amount of …
Orchard: Differentially private analytics at scale
This paper presents Orchard, a system that can answer queries about sensitive data that is
held by millions of user devices, with strong differential privacy guarantees. Orchard …
held by millions of user devices, with strong differential privacy guarantees. Orchard …
Dp-finder: Finding differential privacy violations by sampling and optimization
We present DP-Finder, a novel approach and system that automatically derives lower
bounds on the differential privacy enforced by algorithms. Lower bounds are practically …
bounds on the differential privacy enforced by algorithms. Lower bounds are practically …
Mycelium: Large-scale distributed graph queries with differential privacy
This paper introduces Mycelium, the first system to process differentially private queries over
large graphs that are distributed across millions of user devices. Such graphs occur, for …
large graphs that are distributed across millions of user devices. Such graphs occur, for …
Checkdp: An automated and integrated approach for proving differential privacy or finding precise counterexamples
We propose CheckDP, an automated and integrated approach for proving or disproving
claims that a mechanism is differentially private. CheckDP can find counterexamples for …
claims that a mechanism is differentially private. CheckDP can find counterexamples for …
Guidelines for implementing and auditing differentially private systems
Differential privacy is an information theoretic constraint on algorithms and code. It provides
quantification of privacy leakage and formal privacy guarantees that are currently …
quantification of privacy leakage and formal privacy guarantees that are currently …
Symbolic execution for randomized programs
Z Susag, S Lahiri, J Hsu, S Roy - Proceedings of the ACM on …, 2022 - dl.acm.org
We propose a symbolic execution method for programs that can draw random samples. In
contrast to existing work, our method can verify randomized programs with unknown inputs …
contrast to existing work, our method can verify randomized programs with unknown inputs …