Improving (a,f)-Byzantine resilience in federated learning via layerwise aggregation and cosine distance García-Márquez, Mario Rodríguez-Barroso, Nuria Luzón García, María Victoria Herrera, Francisco P. Federated Learning Robust aggregation Machine learning This research results of the Strategic Project IAFER-Cib (C074/23), as a result of the collaboration agreement signed between the National Institute of Cybersecurity (INCIBE) and the University of Granada. This initiative is carried out within the framework of the Recovery, Transformation, and Resilience Plan funds, financed by the European Union (Next Generation). Funding for open access charge: Universidad de Granada / CBUA The rapid development of artificial intelligence systems has amplified societal concerns regarding their usage, necessitating regulatory frameworks that encompass data privacy. Federated learning (FL) is proposed as a potential solution to the challenges of data privacy in distributed machine learning by enabling collaborative model training without data sharing. However, FL systems remain vulnerable to Byzantine attacks, where malicious nodes contribute corrupted model updates. Although Byzantine resilient rules have emerged as a widely adopted robust aggregation algorithm to mitigate these attacks, their effectiveness drops significantly in high-dimensional parameter spaces, sometimes leading to poor-performing models. This paper introduces Layerwise Cosine Aggregation, a novel aggregation scheme designed to enhance the robustness of these rules in such high-dimensional settings while preserving computational efficiency. A theoretical analysis is presented, demonstrating the superior robustness of the proposed Layerwise Cosine Aggregation compared to the original robust aggregation rules. Empirical evaluation in diverse image classification datasets, under varying data distributions and Byzantine attack scenarios, consistently demonstrates the improved performance of Layerwise Cosine Aggregation, achieving up to a 16% increase in model accuracy. 2025-11-20T12:02:14Z 2025-11-20T12:02:14Z 2025-07-03 journal article García-Márquez, M., Rodríguez-Barroso, N., Luzón, M. V., & Herrera, F. (2025). Improving (α, f)-Byzantine resilience in federated learning via layerwise aggregation and cosine distance. Knowledge-Based Systems, 114004. https://doi.org/10.1016/j.knosys.2025.114004 https://hdl.handle.net/10481/108143 10.1016/j.knosys.2025.114004 eng http://creativecommons.org/licenses/by-nc-nd/4.0/ open access Attribution-NonCommercial-NoDerivatives 4.0 Internacional Elsevier