Search Results for author:
Found 4 papers, 0 papers with code
DP-SGD for non-decomposable objective functions
To overcome this issue, we develop a new DP-SGD variant for similarity based loss functions -- in particular the commonly used contrastive loss -- that manipulates gradients of the objective function in a novel way to obtain a senstivity of the summed gradient that is $O(1)$ for batch size $n$.
Easy Differentially Private Linear Regression
In this paper, we study an algorithm which uses the exponential mechanism to select a model with high Tukey depth from a collection of non-private regression models.
Fast Dimension Independent Private AdaGrad on Publicly Estimated Subspaces
In particular, we show that if the gradients lie in a known constant rank subspace, and assuming algorithmic access to an envelope which bounds decaying sensitivity, one can achieve faster convergence to an excess empirical risk of $\tilde O(1/\epsilon n)$, where $\epsilon$ is the privacy budget and $n$ the number of samples.
Federating Recommendations Using Differentially Private Prototypes
However, in domains that demand protection of personally sensitive data, such as medicine or banking, how can we learn such a model without accessing the sensitive data, and without inadvertently leaking private information?
