Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules
This work is significant for computational chemists and materials scientists working with highly reactive or non-equilibrium molecular systems, providing improved speed and accuracy for property prediction.
This paper addresses the challenge of predicting properties of molecules far from equilibrium by proposing DimeNet++, a model that is 8x faster and 10% more accurate than DimeNet on the QM9 equilibrium dataset. It also introduces the COLL dataset for highly reactive molecules and explores uncertainty quantification for non-equilibrium structures.
Many important tasks in chemistry revolve around molecules during reactions. This requires predictions far from the equilibrium, while most recent work in machine learning for molecules has been focused on equilibrium or near-equilibrium states. In this paper we aim to extend this scope in three ways. First, we propose the DimeNet++ model, which is 8x faster and 10% more accurate than the original DimeNet on the QM9 benchmark of equilibrium molecules. Second, we validate DimeNet++ on highly reactive molecules by developing the challenging COLL dataset, which contains distorted configurations of small molecules during collisions. Finally, we investigate ensembling and mean-variance estimation for uncertainty quantification with the goal of accelerating the exploration of the vast space of non-equilibrium structures. Our DimeNet++ implementation as well as the COLL dataset are available online.