F-KANs: Federated Kolmogorov-Arnold Networks

TL;DR

This paper introduces Federated Kolmogorov-Arnold Networks (F-KANs), a novel federated learning approach that leverages adaptive KANs for improved privacy-preserving classification, outperforming traditional federated MLPs in multiple metrics.

Contribution

The paper proposes F-KANs, integrating Kolmogorov-Arnold Networks into federated learning to enhance classification accuracy and privacy preservation over existing methods.

Findings

F-KANs outperform federated MLPs in accuracy, precision, recall, and F1 score.

F-KANs demonstrate greater stability and efficiency in federated classification tasks.

The approach advances privacy-preserving predictive analytics.

Abstract

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.