AntiFLipper: A Secure and Efficient Defense Against Label-Flipping Attacks in Federated Learning
Aashnan Rahman, Abid Hasan, Sherajul Arifin, Faisal Haque Bappy, Tahrim Hossain, Tariqul Islam, Abu Raihan Mostofa Kamal, Md. Azam Hossain
TL;DR
AntiFLipper tackles label-flipping attacks in federated learning by decentralizing detection and using a trust-based weighted aggregation that shifts computation to clients. It combines local accuracy evaluation, dynamic trust inference, behavior profiling, anomaly-guided filtering, and weighted aggregation to detect and exclude malicious participants with minimal server burden. The approach achieves accuracy comparable to state-of-the-art defenses while significantly reducing aggregation time and maintaining robustness under non-IID distributions and dynamic attack scenarios. This has practical significance for privacy-preserving FL in resource-constrained environments, enabling scalable, secure collaboration without centralized data access or heavy server-side computation.
Abstract
Federated learning (FL) enables privacy-preserving model training by keeping data decentralized. However, it remains vulnerable to label-flipping attacks, where malicious clients manipulate labels to poison the global model. Despite their simplicity, these attacks can severely degrade model performance, and defending against them remains challenging. We introduce AntiFLipper, a novel and computationally efficient defense against multi-class label-flipping attacks in FL. Unlike existing methods that ensure security at the cost of high computational overhead, AntiFLipper employs a novel client-side detection strategy, significantly reducing the central server's burden during aggregation. Comprehensive empirical evaluations across multiple datasets under different distributions demonstrate that AntiFLipper achieves accuracy comparable to state-of-the-art defenses while requiring substantially fewer computational resources in server side. By balancing security and efficiency, AntiFLipper addresses a critical gap in existing defenses, making it particularly suitable for resource-constrained FL deployments where both model integrity and operational efficiency are essential.