Showing 1–1 of 1 results for author: Margraf, J T

Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules

Authors: Johannes Gasteiger, Shankari Giri, Johannes T. Margraf, Stephan Günnemann

Abstract: 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… ▽ More 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. △ Less

Submitted 5 April, 2022; v1 submitted 28 November, 2020; originally announced November 2020.

Comments: Published at the Machine Learning for Molecules Workshop at NeurIPS 2020. Author name changed from Johannes Klicpera to Johannes Gasteiger