Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?

TL;DR

This paper investigates the impact of heterophily on GNNs for node classification, proposing new metrics and an adaptive framework that improve performance on real-world tasks by addressing harmful heterophily.

Contribution

It introduces novel similarity-based metrics for heterophily, and proposes the Adaptive Channel Mixing framework to effectively handle harmful heterophily in GNNs.

Findings

ACM framework improves node classification accuracy.

New metrics outperform traditional homophily metrics.

Achieves state-of-the-art results on multiple datasets.

Abstract

Graph Neural Networks (GNNs) extend basic Neural Networks (NNs) by using the graph structures based on the relational inductive bias (homophily assumption). Though GNNs are believed to outperform NNs in real-world tasks, performance advantages of GNNs over graph-agnostic NNs seem not generally satisfactory. Heterophily has been considered as a main cause and numerous works have been put forward to address it. In this paper, we first show that not all cases of heterophily are harmful for GNNs with aggregation operation. Then, we propose new metrics based on a similarity matrix which considers the influence of both graph structure and input features on GNNs. The metrics demonstrate advantages over the commonly used homophily metrics by tests on synthetic graphs. From the metrics and the observations, we find some cases of harmful heterophily can be addressed by diversification operation. With this fact and knowledge of filterbanks, we propose the Adaptive Channel Mixing (ACM) framework to adaptively exploit aggregation, diversification and identity channels in each GNN layer to address harmful heterophily. We validate the ACM-augmented baselines with 10 real-world node classification tasks. They consistently achieve significant performance gain and exceed the state-of-the-art GNNs on most of the tasks without incurring significant computational burden.