Gradient Coupling Effect of Poisoning Attacks in Federated Learning

Simple item page
dc.contributor.authorWei, Wenqi
dc.contributor.authorLiu, Ling
dc.date.accessioned2023-12-26T18:54:46Z
dc.date.available2023-12-26T18:54:46Z
dc.date.issued2024-01-03
dc.identifier.doihttps://doi.org/10.24251/HICSS.2024.913
dc.identifier.isbn978-0-9981331-7-1
dc.identifier.otherffef8d40-09d4-4b68-b8d0-50ed46c08c01
dc.identifier.urihttps://hdl.handle.net/10125/107299
dc.language.isoeng
dc.relation.ispartofProceedings of the 57th Hawaii International Conference on System Sciences
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectMachine Learning and AI: Cybersecurity and Threat Hunting
dc.subjectfederated learning
dc.subjectpoisoning attacks
dc.subjectsecurity analysis
dc.titleGradient Coupling Effect of Poisoning Attacks in Federated Learning
dc.typeConference Paper
dc.type.dcmiText
dcterms.abstractPoisoning Attack is a dominating threat in distributed learning, where the mediator has limited control over the distributed client contributing to the joint model. In this paper, we present a comprehensive study on the coupling effect of poisoning attacks from three perspectives. First, we identify the theoretical foundation of the weak coupling phenomenon of gradient eigenvalues when under the poisoning attack. Second, we analyze the behavior of gradient coupling under four scenarios: adaptive attacker, skewed client selection, Non-IID data distribution, and different gradient window sizes. We study when the weak coupling effect would fail as the attack indicator. Last, we examine the coupling effect by revisiting several existing poisoning mitigation approaches. Through formal analysis and extensive empirical evidence, we show under what conditions the weak coupling effect of poisoning attacks can serve as forensic evidence for attack mitigation in federated learning and how it interacts with the existing defenses.
dcterms.extent10 pages
prism.startingpage7602

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
0742.pdf
Size:
1.67 MB
Format:
Adobe Portable Document Format
Download

Collections

Machine Learning and AI: Cybersecurity and Threat Hunting