Confidentiality protection in the 2020 US Census of Population and Housing
In an era where external data and computational capabilities far exceed statistical agencies'
own resources and capabilities, they face the renewed challenge of protecting the …
own resources and capabilities, they face the renewed challenge of protecting the …
Crypten: Secure multi-party computation meets machine learning
Secure multi-party computation (MPC) allows parties to perform computations on data while
keeping that data private. This capability has great potential for machine-learning …
keeping that data private. This capability has great potential for machine-learning …
Are we there yet? timing and floating-point attacks on differential privacy systems
Differential privacy is a de facto privacy framework that has seen adoption in practice via a
number of mature software platforms. Implementation of differentially private (DP) …
number of mature software platforms. Implementation of differentially private (DP) …
The distributed discrete gaussian mechanism for federated learning with secure aggregation
We consider training models on private data that are distributed across user devices. To
ensure privacy, we add on-device noise and use secure aggregation so that only the noisy …
ensure privacy, we add on-device noise and use secure aggregation so that only the noisy …
The skellam mechanism for differentially private federated learning
We introduce the multi-dimensional Skellam mechanism, a discrete differential privacy
mechanism based on the difference of two independent Poisson random variables. To …
mechanism based on the difference of two independent Poisson random variables. To …
Practical and private (deep) learning without sampling or shuffling
We consider training models with differential privacy (DP) using mini-batch gradients. The
existing state-of-the-art, Differentially Private Stochastic Gradient Descent (DP-SGD) …
existing state-of-the-art, Differentially Private Stochastic Gradient Descent (DP-SGD) …
[PDF][PDF] The 2020 census disclosure avoidance system topdown algorithm
JM Abowd, R Ashmead… - Harvard Data …, 2022 - assets.pubpub.org
ABSTRACT The Census TopDown Algorithm (TDA) is a disclosure avoidance system using
differential privacy for privacy-loss accounting. The algorithm ingests the final, edited version …
differential privacy for privacy-loss accounting. The algorithm ingests the final, edited version …
Hiding among the clones: A simple and nearly optimal analysis of privacy amplification by shuffling
Recent work of Erlingsson, Feldman, Mironov, Raghunathan, Talwar, and Thakurta 1
demonstrates that random shuffling amplifies differential privacy guarantees of locally …
demonstrates that random shuffling amplifies differential privacy guarantees of locally …
Hyperparameter tuning with renyi differential privacy
N Papernot, T Steinke - arXiv preprint arXiv:2110.03620, 2021 - arxiv.org
For many differentially private algorithms, such as the prominent noisy stochastic gradient
descent (DP-SGD), the analysis needed to bound the privacy leakage of a single training …
descent (DP-SGD), the analysis needed to bound the privacy leakage of a single training …
The fundamental price of secure aggregation in differentially private federated learning
We consider the problem of training a $ d $ dimensional model with distributed differential
privacy (DP) where secure aggregation (SecAgg) is used to ensure that the server only sees …
privacy (DP) where secure aggregation (SecAgg) is used to ensure that the server only sees …