aiqbench analysis
SketchGuard introduces a novel "commit-then-sketch" protocol to secure decentralized federated learning against adaptive null-space attacks, addressing critical scalability issues in Byzantine-robust systems. By decpling screening from aggregation using compact Count Sketch domains, the method significantly reduces communication overhead while maintaining robustness across diverse network topologies and data heterogeneity. This research advances the theoretical foundations of privacy-preserving machine learning by proving convergence guarantees for both convex and non-convex settings in a decentralized environment.
SketchGuard: Scaling Byzantine-Robust Decentralized Federated Learning via Sketch-Based Screening
arXiv:2510.07922v5 Announce Type: replace Abstract: Byzantine-robust decentralized federated learning (DFL) protects peer-to-peer training from malicious clients. The dominant defenses rely on similarity-based filtering, in which each client exchanges full model vectors with every neighbor before any filtering decision; this communication grows with the model dimension and scales poorly as models grow. We propose SketchGuard, which decouples screening from aggregation: clients screen neighbors in a compact Count Sketch domain and fetch full models only from those that pass the screen. We show this idea is insecure when implemented naively. Because the sketch is a fixed, publicly known linear map, an adaptive adversary can hide an arbitrarily large perturbation in its null space, so the poisoned model passes both the sketch-domain filter and the re-sketch verification. We prove this vulnerability and close it with commit-then-sketch, a one-message protocol that draws the sketch seed only after models are committed, restoring the oblivious setting in which Count Sketch provably preserves screening decisions. We then establish convergence in strongly convex and non-convex settings, with explicit dependence on network connectivity and data heterogeneity. Empirically, secured SketchGuard matches state-of-the-art full-precision robustness, up to a small threshold inflation, across six attacks including the adaptive null-space attack, a range of network topologies and heterogeneity settings, and a decentralized fine-tuning task on an 11-million-parameter language model, while reducing per-neighbor screening communication to a size independent of the model dimension.
Abstract:Byzantine-robust decentralized federated learning (DFL) protects peer-to-peer training from malicious clients. The dominant defenses rely on similarity-based filtering, in which each client exchanges full model vectors with every neighbor before any filtering decision; this communication grows with the model dimension and scales poorly as models grow. We propose SketchGuard, which decouples screening from aggregation: clients screen neighbors in a compact Count Sketch domain and fetch full models only from those that pass the screen. We show this idea is insecure when implemented naively. Because the sketch is a fixed, publicly known linear map, an adaptive adversary can hide an arbitrarily large perturbation in its null space, so the poisoned model passes both the sketch-domain filter and the re-sketch verification. We prove this vulnerability and close it with commit-then-sketch, a one-message protocol that draws the sketch seed only after models are committed, restoring the oblivious setting in which Count Sketch provably preserves screening decisions. We then establish convergence in strongly convex and non-convex settings, with explicit dependence on network connectivity and data heterogeneity. Empirically, secured SketchGuard matches state-of-the-art full-precision robustness, up to a small threshold inflation, across six attacks including the adaptive null-space attack, a range of network topologies and heterogeneity settings, and a decentralized fine-tuning task on an 11-million-parameter language model, while reducing per-neighbor screening communication to a size independent of the model dimension.
Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
arXiv:2510.07922 [cs.LG]
(or arXiv:2510.07922v5 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2510.07922 Focus to learn more arXiv-issued DOI via DataCite
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 · 322 words
Related coverage
FUSE: Fairness-aware Update Screening against Byzantine Exploitation in $q$-Fair Federated Learning
arXiv · Oct 15, 2026
Robust Decentralized Personalized Federated Learning via Prediction-Constrained Neighborhood Collaboration
arXiv · Sep 9, 2026
Robust Decentralized Federated Distillation via Multi-Modality Knowledge Collaboration
arXiv · Sep 9, 2026
News
Quantum Computing Report · Jul 25, 2026