English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Inserting a backdoor into the joint model in federated learning (FL) is arecent threat raising concerns. Existing studies mostly focus on developingeffective countermeasures against this threat, assuming that backdoored localmodels, if any, somehow reveal themselves by anomalies in their gradients.However, this assumption needs to be elaborated by identifying specificallywhich gradients are more likely to indicate an anomaly to what extent underwhich conditions. This is an important issue given that neural network modelsusually have huge parametric space and consist of a large number of weights. Inthis study, we make a deep gradient-level analysis on the expected variationsin model gradients under several backdoor attack scenarios against FL. Our mainnovel finding is that backdoor-induced anomalies in local model updates(weights or gradients) appear in the final layer bias weights of the maliciouslocal models. We support and validate our findings by both theoretical andexperimental analysis in various FL settings. We also investigate the impact ofthe number of malicious clients, learning rate, and malicious data rate on theobserved anomaly. Our implementation is publicly available\footnote{\url{https://github.com/ArcelikAcikKaynak/Federated_Learning.git}}.

Anomaly Localization in Model Gradients Under Backdoor Attacks Against Federated Learning