We consider multi-party protocols for classification that are motivated ...
A canonical noise distribution (CND) is an additive mechanism designed t...
In this paper, we consider the framework of privacy amplification via
it...
Characterizing the privacy degradation over compositions, i.e., privacy
...
In this rejoinder, we aim to address two broad issues that cover most
co...
Perhaps the single most important use case for differential privacy is t...
Datasets containing sensitive information are often sequentially analyze...
Deep learning models are often trained on datasets that contain sensitiv...
Composition is one of the most important properties of differential priv...
We study a game between two firms in which each provide a service based ...
Differential privacy has seen remarkable success as a rigorous and pract...
We study an online linear classification problem, in which the data is
g...