Neural Message Passing for Quantum Chemistry

TL;DR

This paper introduces a unified framework called Message Passing Neural Networks (MPNNs) for molecular property prediction, demonstrating state-of-the-art results and suggesting future research directions for larger molecules and more precise data.

Contribution

The paper reformulates existing neural message passing models into a single framework and explores novel variations, achieving state-of-the-art performance on chemical prediction benchmarks.

Findings

Achieved state-of-the-art results on molecular property prediction benchmarks.

Unified various message passing models into the MPNN framework.

Indicated future research should focus on larger molecules or more accurate labels.

Abstract

Supervised learning on molecules has incredible potential to be useful in chemistry, drug discovery, and materials science. Luckily, several promising and closely related neural network models invariant to molecular symmetries have already been described in the literature. These models learn a message passing algorithm and aggregation procedure to compute a function of their entire input graph. At this point, the next step is to find a particularly effective variant of this general approach and apply it to chemical prediction benchmarks until we either solve them or reach the limits of the approach. In this paper, we reformulate existing models into a single common framework we call Message Passing Neural Networks (MPNNs) and explore additional novel variations within this framework. Using MPNNs we demonstrate state of the art results on an important molecular property prediction benchmark; these results are strong enough that we believe future work should focus on datasets with larger molecules or more accurate ground truth labels.