| dc.description.abstract | Federated Learning (FL) permits different parties to collaboratively train a global model without disclosing their respective local labels. A crucial step of FL, that of aggregating local models to produce the global one, shares many similarities with public decision-making, and elections in particular. In that context, a major weakness of FL, namely its vulnerability to poisoning attacks, can be interpreted as a consequence of the \emph{one person one vote} (henceforth \emph{1p1v}) principle that underpins most contemporary aggregation rules.
In this paper, we introduce \textsc{FedQV}, a novel aggregation algorithm built upon the \emph{quadratic voting} scheme, recently proposed as a better alternative to \emph{1p1v}-based elections. Our theoretical analysis establishes that \textsc{FedQV} is a truthful mechanism in which bidding according to one's true valuation is a dominant strategy that achieves a convergence rate matching that of state-of-the-art methods. Furthermore, our empirical analysis using multiple real-world datasets validates the superior performance of \textsc{FedQV} against poisoning attacks. It also shows that combining \textsc{FedQV} with unequal voting ``budgets'' according to a reputation score increases its performance benefits even further. Finally, we show that \textsc{FedQV} can be easily combined with Byzantine-robust privacy-preserving mechanisms to enhance its robustness against both poisoning and privacy attacks. | es |
