New FedIoC Framework Detects Cyberattack Campaigns Using Federated Learning

Researchers have developed FedIoC, a new framework designed to detect coordinated cyberattack campaigns across multiple organizations. This system utilizes Federated Learning to train threat detectors on local data without sharing sensitive information. FedIoC encodes local threat indicators into gradient updates, allowing the central server to cluster these updates based on cosine similarity and identify global campaign patterns. Evaluations on public benchmarks demonstrate FedIoC's ability to recover cross-organizational attack cohorts by analyzing gradient geometry, even when clients have disjoint indicator sets. AI

IMPACT This research could enhance cybersecurity defenses by enabling collaborative threat detection without compromising data privacy.

RANK_REASON The item is a research paper published on arXiv detailing a new technical framework. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• Federated Learning
• FedIoC
• Gotit.pub
• Hugging Face
• IArxiv Recommender
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →