Advances and open problems in federated learning

P Kairouz, HB McMahan, B Avent… - … and trends® in …, 2021 - nowpublishers.com
Federated learning (FL) is a machine learning setting where many clients (eg, mobile
devices or whole organizations) collaboratively train a model under the orchestration of a …

The distributed discrete gaussian mechanism for federated learning with secure aggregation

P Kairouz, Z Liu, T Steinke - International Conference on …, 2021 - proceedings.mlr.press
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 …

The fundamental price of secure aggregation in differentially private federated learning

WN Chen, CAC Choo, P Kairouz… - … on Machine Learning, 2022 - proceedings.mlr.press
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 …

Dp-cryptography: marrying differential privacy and cryptography in emerging applications

S Wagh, X He, A Machanavajjhala… - Communications of the …, 2021 - dl.acm.org
DP-cryptography: marrying differential privacy and cryptography in emerging applications
Page 1 84 COMMUNICATIONS OF THE ACM | FEBRUARY 2021 | VOL. 64 | NO. 2 review …

Scenario-based Adaptations of Differential Privacy: A Technical Survey

Y Zhao, JT Du, J Chen - ACM Computing Surveys, 2024 - dl.acm.org
Differential privacy has been a de facto privacy standard in defining privacy and handling
privacy preservation. It has had great success in scenarios of local data privacy and …

Private summation in the multi-message shuffle model

B Balle, J Bell, A Gascón, K Nissim - Proceedings of the 2020 ACM …, 2020 - dl.acm.org
The shuffle model of differential privacy (Erlingsson et al. SODA 2019; Cheu et al.
EUROCRYPT 2019) and its close relative encode-shuffle-analyze (Bittau et al. SOSP 2017) …

Stronger privacy amplification by shuffling for rényi and approximate differential privacy

V Feldman, A McMillan, K Talwar - Proceedings of the 2023 Annual ACM …, 2023 - SIAM
The shuffle model of differential privacy has gained significant interest as an intermediate
trust model between the standard local and central models [18, 12]. A key result in this …

On the power of multiple anonymous messages: Frequency estimation and selection in the shuffle model of differential privacy

B Ghazi, N Golowich, R Kumar, R Pagh… - … Conference on the …, 2021 - Springer
It is well-known that general secure multi-party computation can in principle be applied to
implement differentially private mechanisms over distributed data with utility matching the …

User-level differentially private learning via correlated sampling

B Ghazi, R Kumar… - Advances in Neural …, 2021 - proceedings.neurips.cc
Most works in learning with differential privacy (DP) have focused on the setting where each
user has a single sample. In this work, we consider the setting where each user holds $ m …

Scalable and differentially private distributed aggregation in the shuffled model

B Ghazi, R Pagh, A Velingker - arxiv preprint arxiv:1906.08320, 2019 - arxiv.org
Federated learning promises to make machine learning feasible on distributed, private
datasets by implementing gradient descent using secure aggregation methods. The idea is …