Catch Causal Signals from Edges for Label Imbalance in Graph Classification

TL;DR

This paper introduces a causal attention mechanism that leverages edge features to better detect causal signals in graphs, improving classification performance under label imbalance conditions.

Contribution

It enhances causal detection in graphs by effectively utilizing edge information, leading to better disentanglement of causal subgraphs and improved classification results.

Findings

Improved performance on PTC, Tox21, and ogbg-molhiv datasets.

Edge features significantly enhance causal signal detection.

Addresses label imbalance in graph classification.

Abstract

Despite significant advancements in causal research on graphs and its application to cracking label imbalance, the role of edge features in detecting the causal effects within graphs has been largely overlooked, leaving existing methods with untapped potential for further performance gains. In this paper, we enhance the causal attention mechanism through effectively leveraging edge information to disentangle the causal subgraph from the original graph, as well as further utilizing edge features to reshape graph representations. Capturing more comprehensive causal signals, our design leads to improved performance on graph classification tasks with label imbalance issues. We evaluate our approach on real-word datasets PTC, Tox21, and ogbg-molhiv, observing improvements over baselines. Overall, we highlight the importance of edge features in graph causal detection and provide a promising direction for addressing label imbalance challenges in graph-level tasks. The model implementation details and the codes are available on https://github.com/fengrui-z/ECAL