Mohammad Naseri (University College London), Jamie Hayes (DeepMind), Emiliano De Cristofaro (University College London & Alan Turing Institute)

Federated Learning (FL) allows multiple participants to train machine learning models collaboratively by keeping their datasets local while only exchanging model updates. Alas, this is not necessarily free from privacy and robustness vulnerabilities, e.g., via membership, property, and backdoor attacks. This paper investigates whether and to what extent one can use differential Privacy (DP) to protect both privacy and robustness in FL. To this end, we present a first-of-its-kind evaluation of Local and Central Differential Privacy (LDP/CDP) techniques in FL, assessing their feasibility and effectiveness.

Our experiments show that both DP variants do defend against backdoor attacks, albeit with varying levels of protection-utility trade-offs, but anyway more effectively than other robustness defenses. DP also mitigates white-box membership inference attacks in FL, and our work is the first to show it empirically. Neither LDP nor CDP, however, defend against property inference. Overall, our work provides a comprehensive, re-usable measurement methodology to quantify the trade-offs between robustness/privacy and utility in differentially private FL.

View More Papers

Fuzzing Configurations of Program Options

Zenong Zhang (University of Texas at Dallas), George Klees (University of Maryland), Eric Wang (Poolesville High School), Michael Hicks (University of Maryland), Shiyi Wei (University of Texas at Dallas)

Read More

Demo #7: A Simulator for Cooperative and Automated Driving...

Mohammed Lamine Bouchouia (Telecom Paris - Institut Polytechnique de Paris), Jean-Philippe Monteuuis (Qualcomm Technologies Inc), Houda Labiod (Telecom Paris - Institut Polytechnique de Paris), Ons Jelassi (Telecom Paris - Institut Polytechnique de Paris), Wafa Ben Jaballah (Thales) and Jonathan Petit (Qualcomm Technologies Inc)

Read More

On Utility and Privacy in Synthetic Genomic Data

Bristena Oprisanu (UCL), Georgi Ganev (UCL & Hazy), Emiliano De Cristofaro (UCL)

Read More

Generation of CAN-based Wheel Lockup Attacks on the Dynamics...

Alireza Mohammadi (University of Michigan-Dearborn), Hafiz Malik (University of Michigan-Dearborn) and Masoud Abbaszadeh (GE Global Research)

Read More