MetaGMT: Improving Actionable Interpretability of Graph Multilinear Networks via Meta-Learning Filtration

TL;DR

MetaGMT introduces a meta-learning framework that significantly enhances the explanation quality and robustness of Graph Multi-linear Networks, making them more trustworthy for high-stakes applications.

Contribution

It proposes a novel bi-level optimization approach to improve explanation fidelity and robustness of GNN explanations, addressing spurious correlations and trust issues.

Findings

Significantly improves explanation AUC-ROC and Precision@K across benchmarks.

Maintains competitive classification accuracy while enhancing explanation faithfulness.

Increases explanation ROC by up to 8% on SP-Motif 0.5.

Abstract

The growing adoption of Graph Neural Networks (GNNs) in high-stakes domains like healthcare and finance demands reliable explanations of their decision-making processes. While inherently interpretable GNN architectures like Graph Multi-linear Networks (GMT) have emerged, they remain vulnerable to generating explanations based on spurious correlations, potentially undermining trust in critical applications. We present MetaGMT, a meta-learning framework that enhances explanation fidelity through a novel bi-level optimization approach. We demonstrate that MetaGMT significantly improves both explanation quality (AUC-ROC, Precision@K) and robustness to spurious patterns, across BA-2Motifs, MUTAG, and SP-Motif benchmarks. Our approach maintains competitive classification accuracy while producing more faithful explanations (with an increase up to 8% of Explanation ROC on SP-Motif 0.5) compared to baseline methods. These advancements in interpretability could enable safer deployment of GNNs in sensitive domains by (1) facilitating model debugging through more reliable explanations, (2) supporting targeted retraining when biases are identified, and (3) enabling meaningful human oversight. By addressing the critical challenge of explanation reliability, our work contributes to building more trustworthy and actionable GNN systems for real-world applications.