Catch Causal Signals from Edges for Label Imbalance in Graph Classification

2025-01-03

AI Summary

Graph classification often suffers from label imbalance, leading to biased model performance and unfair predictions. Method: This paper pioneers a systematic investigation into the critical role of edge features in identifying causal signals, proposing an edge-driven causal subgraph disentanglement and representation reformation framework. It introduces a causal attention mechanism that jointly encodes edge features and disentangles subgraphs to explicitly isolate causally relevant structures; subsequently, a GNN-based reconstruction module debiases graph representations, overcoming the limitations of conventional node-centric causal modeling. Contribution/Results: Extensive experiments on PTC, Tox21, and ogbg-molhiv demonstrate that the method significantly outperforms state-of-the-art baselines, effectively mitigating performance degradation induced by label imbalance. It establishes a novel paradigm for graph-level causal learning and fair classification.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Social Networks and Social Media: Fairness and bias in social network and social media analysis

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

Problem

• Graph Classification
• Imbalanced Labels
• Edge Utilization

Innovation

• Causal Attention Mechanism
• Imbalanced Label Handling
• Graph Representation Redesign