In this paper, we present an innovative federated learning (FL) approach that utilizes Kolmogorov-Arnold Networks (KANs) for classification tasks. By utilizing the adaptive activation capabilities of KANs in a federated framework, we aim to improve classification capabilities while preserving privacy. The study evaluates the performance of federated KANs (F- KANs) compared to traditional Multi-Layer Perceptrons (MLPs) on classification task. The results show that the F-KANs model significantly outperforms the federated MLP model in terms of accuracy, precision, recall, F1 score and stability, and achieves better performance, paving the way for more efficient and privacy-preserving predictive analytics.
@article{arxiv.2407.20100,
title = {F-KANs: Federated Kolmogorov-Arnold Networks},
author = {Engin Zeydan and Cristian J. Vaca-Rubio and Luis Blanco and Roberto Pereira and Marius Caus and Abdullah Aydeger},
journal= {arXiv preprint arXiv:2407.20100},
year = {2024}
}
Comments
This work has been accepted to 1st International Workshop on Distributed AI for Enhanced Wireless Networks (DAINET'25) in conjunction with IEEE Consumer Communications & Networking Conference 2025. Related Code: https://github.com/ezeydan/F-KANs.git