Enhancing Graph Collaborative Filtering with FourierKAN Feature Transformation

TL;DR

This paper introduces FourierKAN-GCF, a novel graph convolutional framework that improves recommendation accuracy by incorporating Fourier Kolmogorov-Arnold Networks for feature transformation, balancing model expressiveness and training complexity.

Contribution

It proposes FourierKAN-GCF, a new GCN framework using Fourier Kolmogorov-Arnold Networks to enhance feature transformation in graph collaborative filtering models.

Findings

FourierKAN-GCF outperforms existing GCF models on public datasets.

The framework can be integrated into self-supervised models for improved performance.

Enhanced feature transformation reduces training difficulty while boosting recommendation accuracy.

Abstract

Graph Collaborative Filtering (GCF) has emerged as a dominant paradigm in modern recommendation systems, excelling at modeling complex user-item interactions and capturing high-order collaborative signals through graph-structured learning. Most existing GCF models predominantly rely on simplified graph architectures like LightGCN, which strategically remove feature transformation and activation functions from vanilla graph convolution networks. Through systematic analysis, we reveal that feature transformation in message propagation can enhance model representation, though at the cost of increased training difficulty. To this end, we propose FourierKAN-GCF, a novel GCN framework that adopts Fourier Kolmogorov-Arnold Networks as efficient transformation modules within graph propagation layers. This design enhances model representation while decreasing training difficulty. Our FourierKAN-GCF can achieve higher recommendation performance than most widely used GCF backbone models. In addition, it can be integrated into existing advanced self-supervised models as a backbone, replacing their original backbone to achieve enhanced performance. Extensive experiments on three public datasets demonstrate the superiority of FourierKAN-GCF.