The paper entitled “Shielding Federated Learning Systems agains Interence Attacks with ARM TrustZone” was Best Paper Award Runner-Up in Middleware 2022, the 23rd ACM/ IFIP Middleware Conference (7-11 November). The paper is authored by Aguiles Ait Messaoud, Vlad Nitu (INSA Lyon, France), Sonia Ben Mokhtar (LIRIS CNRS, France), and also by Valerio Shiavoni (University of Neuchatel, Switzerland), VEDLIoT partner.
This paper shows how, in the context of federated learning, it is possible protect deep-learning inference workflows running on IoT devices from malicious users launching so-called inference attacks. GradSec, the proposed solution, shields specific deep-learning layers into TrustZone enclaves, i.e. secure hardware areas available in off-the-shelf commodity edge devices.
Congratulations to the authors!
