Enhancing Security in Federated Learning through Adaptive Consensus-Based Model Update Validation

arXiv:2403.04803v1 Announce Type: cross
Abstract: This paper introduces an advanced approach for fortifying Federated Learning (FL) systems against label-flipping attacks. We propose a simplified consensus-based verification process integrated with an adaptive thresholding mechanism. This dynamic thresholding is designed to adjust based on the evolving landscape of model updates, offering a refined layer of anomaly detection that aligns with the real-time needs of distributed learning environments. Our method necessitates a majority consensus among participating clients to validate updates, ensuring that only …

abstract advanced arxiv attacks consensus cs.ai cs.cr cs.dc cs.lg dynamic federated learning landscape paper process security simplified systems thresholding through type update updates validation verification