aiqbench analysis
This paper tackles a specific gap in federated learning: fairness-weighted aggregation (q-FFL) can amplify high-loss clients, which makes it hard for a server to distinguish legitimate amplification from malicious update scaling. The proposed FUSE aggregator uses a small trusted labeled dataset and structural and radial screening tests. The reported gains are benchmark-specific, with a 3.9 to 30.1 point improvement in worst-attack accuracy across four datasets and seven attacks, so readers should check how well the assumptions, such as shared class-conditional label skew, hold in their deployments.
FUSE: Fairness-aware Update Screening against Byzantine Exploitation in $q$-Fair Federated Learning
arXiv:2610.20264v1 Announce Type: new Abstract: Client-level fair federated learning, in the form of $q$-fair federated learning ($q$-FFL), agnostic federated learning, or tilted empirical risk minimization, has emerged as an attractive solution to equalize utility across heterogeneous clients. This paradigm grants high-loss clients greater relative influence, while a central server aggregates their loss-weighted updates into a single shared model. Yet, the Byzantine robustness of $q$-fair federated learning has been largely neglected, leaving the server unable to tell legitimate fairness-induced amplification from malicious update scaling. To address this critical gap, we formalize the problem of Byzantine-robust $q$-fair federated learning under an \emph{update-only interface}, in which the server observes nothing but the fairness-scaled client updates. Our analysis under shared class-conditional label skew shows that fairness scaling preserves the nonnegative class-mixture structure of honest updates, while replacing their unit-sum constraint by a client-dependent radial scale. Moving beyond magnitude-based screening, we show that this warped geometry supplies two complementary tests, one radial and one structural, that a single norm cannot provide. We then introduce FUSE, a structure-aware robust aggregator that uses a small trusted labeled dataset for coefficient recovery, scale and cone-consistency screening, and post-screening step-size control. Across four benchmarks and seven Byzantine attacks, FUSE improves seed-wise worst-attack accuracy over the strongest baseline on each benchmark by 3.9--30.1 percentage points, while largely preserving the influence of the high-loss honest clients that fairness is meant to protect.
Abstract:Client-level fair federated learning, in the form of $q$-fair federated learning ($q$-FFL), agnostic federated learning, or tilted empirical risk minimization, has emerged as an attractive solution to equalize utility across heterogeneous clients. This paradigm grants high-loss clients greater relative influence, while a central server aggregates their loss-weighted updates into a single shared model. Yet, the Byzantine robustness of $q$-fair federated learning has been largely neglected, leaving the server unable to tell legitimate fairness-induced amplification from malicious update scaling. To address this critical gap, we formalize the problem of Byzantine-robust $q$-fair federated learning under an \emph{update-only interface}, in which the server observes nothing but the fairness-scaled client updates. Our analysis under shared class-conditional label skew shows that fairness scaling preserves the nonnegative class-mixture structure of honest updates, while replacing their unit-sum constraint by a client-dependent radial scale. Moving beyond magnitude-based screening, we show that this warped geometry supplies two complementary tests, one radial and one structural, that a single norm cannot provide. We then introduce FUSE, a structure-aware robust aggregator that uses a small trusted labeled dataset for coefficient recovery, scale and cone-consistency screening, and post-screening step-size control. Across four benchmarks and seven Byzantine attacks, FUSE improves seed-wise worst-attack accuracy over the strongest baseline on each benchmark by 3.9--30.1 percentage points, while largely preserving the influence of the high-loss honest clients that fairness is meant to protect.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.20264 [cs.LG]
(or arXiv:2610.20264v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.20264 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 323 words
Related coverage
Inside the suddenly explosive world of AI safety
The Verge · Sep 17, 2026
Construct validity of Claude Opus 4.8's System Card – A commentary
LessWrong · Jun 12, 2026
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
Taboo “equilibrium”: Less confused frames for research on AI bargaining
LessWrong · Jul 31, 2026