Is Graph Convolution Always Beneficial For Every Feature?

ICLR 2025 Conference Submission1933 Authors

Keywords: Graph Neural Networks, Graph Homophily, Topological Feature Selection

TL;DR: We propose a new metric to identify GNN-favored and GNN-disfavored features and use topological feature selection to fuse these features into GNNs, which significantly improves GNNs performance without hyper-parameter tuning.

Abstract: Graph Neural Networks (GNNs) have demonstrated strong capabilities in processing structured data. While traditional GNNs typically treat each feature dimension equally during graph convolution, we raise an important question: *Is the graph convolution operation equally beneficial for each feature?* If not, the convolution operation on certain feature dimensions can possibly lead to harmful effects, even worse than the convolution-free models. Traditional feature selection methods focus on identifying informative features or reducing redundancy, but they are not suitable for structured data since they overlook graph structures. In the context of graphs, few studies have investigated GNN performance concerning node features using feature homophily metrics, which assess feature consistency with graph topology. Unfortunately, these metrics have not effectively aligned with GNN performance or served as reliable guides for feature selection in GNNs. To address these limitations, we introduce a novel metric, Topological Feature Informativeness (TFI), to distinguish between GNN-favored and GNN-disfavored features, where its effectiveness is validated through both theoretical analysis and empirical observations. Based on TFI, we propose a simple yet effective Graph Feature Selection (GFS) method, which processes GNN-favored and GNN-disfavored features separately, using GNNs and non-GNN models. Compared to original GNNs, GFS significantly improves the extraction of useful topological information from each feature with comparable computational costs. Extensive experiments show that after applying GFS to $8$ baseline and state-of-the-art (SOTA) GNN architectures across $10$ datasets, $90$\% of the GFS-augmented cases show significant performance boosts. Furthermore, our proposed TFI metric outperforms other feature selection methods in graphs. These results validate the effectiveness of both GFS and TFI. Additionally, we demonstrate that GFS's improvements are robust to hyperparameter tuning, highlighting its potential as a universal method for enhancing various GNN architectures.

Primary Area: learning on graphs and other geometries & topologies

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 1933