Swarm learning is a decentralized and trustdistributed method of collaborative machine learning, which enables multiple participants to jointly train models for malware, intrusion, and anomaly detection without exchanging raw data, protecting privacy, and facilitating regulatory compliance. Swarm Learning has been widely used in healthcare, finance, and cybersecurity domains, where threats are constantly changing and data is sensitive by the use of blockchain and consensus, which ensures tamper-resistant coordination. In this paper, we present a comprehensive overview of the concept, operation processes, and architectural framework of Swarm Learning, focusing on its application in cybersecurity, including its use in threat detection and anomalies detection, as well as other security purposes. We also highlight notable projects that highlight its practical feasibility. We conclude with a review of the existing literature on topics such as swarm learning, federated learning, and decentralized machine learning techniques, particularly in relation to their application to cybersecurity.
Related links
Details
Title
Swarm Learning: A Comprehensive Overview and Its Potential in Cybersecurity
Publication Details
2026 2nd International Conference on Federated Learning and Intelligent Computing Systems (FLICS), pp.290-295
Resource Type
Conference proceeding
Conference
International Conference on Federated Learning and Intelligent Computing Systems (FLICS), 2nd (Valencia, Spain, 06/09/2026–06/12/2026)