Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials
This work addresses the problem of accurate property prediction for molecules and materials, which is crucial for fields like chemistry and materials science, by proposing an incremental improvement to existing neural message passing models.
The authors tackled the problem of predicting formation energy and other properties of molecules and materials by extending neural message passing models with an edge update network, which allows information exchange to depend on the receiving atom's hidden state. They demonstrated superior prediction accuracy on three datasets (QM9, The Materials Project, OQMD) compared to best published results, and found that using a K-nearest neighbors graph for crystalline structures yields better accuracy than other methods.
Neural message passing on molecular graphs is one of the most promising methods for predicting formation energy and other properties of molecules and materials. In this work we extend the neural message passing model with an edge update network which allows the information exchanged between atoms to depend on the hidden state of the receiving atom. We benchmark the proposed model on three publicly available datasets (QM9, The Materials Project and OQMD) and show that the proposed model yields superior prediction of formation energies and other properties on all three datasets in comparison with the best published results. Furthermore we investigate different methods for constructing the graph used to represent crystalline structures and we find that using a graph based on K-nearest neighbors achieves better prediction accuracy than using maximum distance cutoff or the Voronoi tessellation graph.