Fine-grained Graph Rationalization

2023-12-13

AI Summary

Existing graph rationalization methods predominantly employ coarse-grained, graph-level interventions, limiting their ability to precisely identify fine-grained subgraphs—such as critical nodes and virtual nodes—that decisively influence predictions, thereby undermining interpretability and causal robustness. To address this, we propose the first fine-grained graph rationalization framework supporting node-level and virtual-node-level interventions. Our method models node interactions via self-attention, introduces differentiable subgraph masks, and incorporates counterfactual intervention mechanisms for end-to-end joint optimization. Evaluated on seven real-world graph datasets, it significantly outperforms 13 state-of-the-art baselines. Crucially, it achieves higher predictive accuracy while generating more compact, semantically invariant, and discriminative rationale subgraphs. This advancement substantially enhances both the interpretability and causal credibility of graph neural networks.

Abstract

Rationale discovery is defined as finding a subset of the input data that maximally supports the prediction of downstream tasks. In the context of graph machine learning, graph rationale is defined to locate the critical subgraph in the given graph topology. In contrast to the rationale subgraph, the remaining subgraph is named the environment subgraph. Graph rationalization can enhance the model performance as the mapping between the graph rationale and prediction label is viewed as invariant, by assumption. To ensure the discriminative power of the extracted rationale subgraphs, a key technique named"intervention"is applied whose heart is that given changing environment subgraphs, the semantics from the rationale subgraph is invariant, guaranteeing the correct prediction result. However, most, if not all, of the existing graph rationalization methods develop their intervention strategies on the graph level, which is coarse-grained. In this paper, we propose fine-grained graph rationalization (FIG). Our idea is driven by the self-attention mechanism, which provides rich interactions between input nodes. Based on that, FIG can achieve node-level and virtual node-level intervention. Our experiments involve 7 real-world datasets, and the proposed FIG shows significant performance advantages compared to 13 baseline methods.

Problem

• Coarse-grained Intervention
• Graph Rationalization
• Critical Subgraphs

Innovation

• Fine-Grained Graph Rationalization
• Node-Level Intervention
• Self-Attention Mechanism