warisgill / FedDefender

FedDefender is a novel defense mechanism designed to safeguard Federated Learning from the poisoning attacks (i.e., backdoor attacks).

FedDefender: Backdoor Attack Defense in Federated Learning (Tutorial)

This tutorial is based on a paper accepted at SE4SafeML: Dependability and Trustworthiness of Safety-Critical Systems with Machine Learned Components (Colocated with FSE 2023). The ArXiv version of the manuscript is available here.

For any questions regarding FedDefender's artifact, please direct them to Waris Gill at [e-mail].