activity
20122026
most citedDifferentially private cross-silo federated learning

22 citations · 42 across the 12 of their papers we have counts for

collaborators
Showing stat.MLShow all

8 papers · 1 filter

stat.ML2024

Subsampling is not Magic: Why Large Batch Sizes Work for Differentially Private Stochastic Optimisation

Ossi Räisä, Joonas Jälkö, Antti Honkela

We study how the batch size affects the total gradient variance in differentially private stochastic gradient descent (DP-SGD), seeking a theoretical explanation for the usefulness…

stat.ML2022

DPVIm: Differentially Private Variational Inference Improved

Joonas Jälkö, Lukas Prediger, Antti Honkela +1

Differentially private (DP) release of multidimensional statistics typically considers an aggregate sensitivity, e.g. the vector norm of a high-dimensional vector. However, differe…

stat.ML20211 cited

Locally Differentially Private Bayesian Inference

Tejas Kulkarni, Joonas Jälkö, Samuel Kaski +1

In recent years, local differential privacy (LDP) has emerged as a technique of choice for privacy-preserving data collection in several scenarios when the aggregator is not trustw…

stat.ML20191 cited

Differentially Private Federated Variational Inference

Mrinank Sharma, Michael Hutchinson, Siddharth Swaroop +2

In many real-world applications of machine learning, data are distributed across many clients and cannot leave the devices they are stored on. Furthermore, each client's data, comp…

stat.ML2019

Computing Tight Differential Privacy Guarantees Using FFT

Antti Koskela, Joonas Jälkö, Antti Honkela

Differentially private (DP) machine learning has recently become popular. The privacy loss of DP algorithms is commonly reported using -DP. In this paper, we propo…

stat.ML201911 cited

Differentially Private Markov Chain Monte Carlo

Mikko A. Heikkilä, Joonas Jälkö, Onur Dikmen +1

Recent developments in differentially private (DP) machine learning and DP Bayesian learning have enabled learning under strong privacy guarantees for the training data subjects. I…