DropKAN: Regularizing KANs by masking post-activations

TL;DR

DropKAN introduces a simple regularization technique for Kolmogorov-Arnold Networks by masking activations, which improves their generalization performance compared to standard Dropout.

Contribution

This paper presents DropKAN, a novel regularization method that embeds dropout masks within KAN layers, enhancing their generalization ability.

Findings

DropKAN outperforms standard Dropout in KANs on real datasets.

Embedding masks within KAN layers improves regularization effectiveness.

DropKAN is easy to implement with minimal coding effort.

Abstract

We propose DropKAN (Dropout Kolmogorov-Arnold Networks) a regularization method that prevents co-adaptation of activation function weights in Kolmogorov-Arnold Networks (KANs). DropKAN functions by embedding the drop mask directly within the KAN layer, randomly masking the outputs of some activations within the KANs' computation graph. We show that this simple procedure that require minimal coding effort has a regularizing effect and consistently lead to better generalization of KANs. We analyze the adaptation of the standard Dropout with KANs and demonstrate that Dropout applied to KANs' neurons can lead to unpredictable behavior in the feedforward pass. We carry an empirical study with real world Machine Learning datasets to validate our findings. Our results suggest that DropKAN is consistently a better alternative to using standard Dropout with KANs, and improves the generalization performance of KANs. Our implementation of DropKAN is available at: \url{https://github.com/Ghaith81/dropkan}.