Timezone: »

 
Poster
GemNet: Universal Directional Graph Neural Networks for Molecules
Johannes Gasteiger · Florian Becker · Stephan Günnemann

Fri Dec 10 08:30 AM -- 10:00 AM (PST) @
in Poster Session 8 »

Effectively predicting molecular interactions has the potential to accelerate molecular dynamics by multiple orders of magnitude and thus revolutionize chemical simulations. Graph neural networks (GNNs) have recently shown great successes for this task, overtaking classical methods based on fixed molecular kernels. However, they still appear very limited from a theoretical perspective, since regular GNNs cannot distinguish certain types of graphs. In this work we close this gap between theory and practice. We show that GNNs with directed edge embeddings and two-hop message passing are indeed universal approximators for predictions that are invariant to translation, and equivariant to permutation and rotation. We then leverage these insights and multiple structural improvements to propose the geometric message passing neural network (GemNet). We demonstrate the benefits of the proposed changes in multiple ablation studies. GemNet outperforms previous models on the COLL, MD17, and OC20 datasets by 34%, 41%, and 20%, respectively, and performs especially well on the most challenging molecules. Our implementation is available online.

Keywords: Theory · Deep Learning · Graph Learning

Author Information

Johannes Gasteiger (Technical University of Munich)
Florian Becker (Department of Informatics, Technical University Munich)
Stephan Günnemann (Technical University of Munich)

More from the Same Authors