<ul><li>Graph Neural Networks (GNNs) have shown remarkable success in molecular tasks, yet their interpretability remains challenging.</li><li>To address limitations in traditional model-level explanation methods, an innovative approach called MAGE (Motif-based GNN Explainer) has been introduced.</li><li>MAGE uses motifs as fundamental units for generating explanations and incorporates critical substructures into the explanations.</li><li>The effectiveness of MAGE has been demonstrated through quantitative and qualitative assessments on real-world molecular datasets.</li></ul>

MAGE: Model-Level Graph Neural Networks Explanations via Motif-based Graph Generation

Discover more