Explaining Deep Graph Networks with Molecular Counterfactuals
This addresses the problem of interpretability for deep graph networks in molecular science, though it appears incremental as it builds on existing counterfactual explanation methods.
The paper tackles explainability of deep graph networks for molecule property prediction by generating valid counterfactual compounds with high structural similarity but different predicted properties, showing preliminary results that convey key insights to non-ML experts.
We present a novel approach to tackle explainability of deep graph networks in the context of molecule property prediction tasks, named MEG (Molecular Explanation Generator). We generate informative counterfactual explanations for a specific prediction under the form of (valid) compounds with high structural similarity and different predicted properties. We discuss preliminary results showing how the model can convey non-ML experts with key insights into the learning model focus in the neighborhood of a molecule.