Journal article

The Cost of Privacy in Asynchronous Differentially-Private Machine Learning

Farhad Farokhi, Nan Wu, David Smith, Mohamed Ali Kaafar

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | Published : 2021

Abstract

We consider training machine learning models using data located on multiple private and geographically-scattered servers with different privacy settings. Due to the distributed nature of the data, communicating with all collaborating private data owners simultaneously may prove challenging or altogether impossible. We consider differentially-private asynchronous algorithms for collaboratively training machine-learning models on multiple private datasets. The asynchronous nature of the algorithms implies that a central learner interacts with the private data owners one-on-one whenever they are available for communication without needing to aggregate query responses to construct gradients of t..

View full abstract

Grants

Funding Acknowledgements

This research has been supported by the Optus Macquarie University Cyber Security Hub and by the Next Generation Technology Funding from the Defense Science and Technology Group as part of the DPAIP Project.