Fine-grained Graph Rationalization
This work addresses a bottleneck in graph machine learning for researchers and practitioners by improving the precision of rationale discovery, though it is incremental as it builds on existing graph rationalization frameworks.
The paper tackles the problem of coarse-grained intervention strategies in graph rationalization methods by proposing Fine-grained Graph Rationalization (FIG), which uses self-attention to enable node-level and virtual node-level interventions, achieving significant performance advantages over 13 baselines on 7 real-world datasets.
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.