Combining Stochastic Explainers and Subgraph Neural Networks can Increase Expressivity and Interpretability

Open in new window