GemNet: Universal Directional Graph Neural Networks for Molecules

Johannes Klicpera, Florian Becker, Stephan Günnemann

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

220 Zitate (Scopus)

Abstract

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.1

OriginalspracheEnglisch
TitelAdvances in Neural Information Processing Systems 34 - 35th Conference on Neural Information Processing Systems, NeurIPS 2021
Redakteure/-innenMarc'Aurelio Ranzato, Alina Beygelzimer, Yann Dauphin, Percy S. Liang, Jenn Wortman Vaughan
Herausgeber (Verlag)Neural information processing systems foundation
Seiten6790-6802
Seitenumfang13
ISBN (elektronisch)9781713845393
PublikationsstatusVeröffentlicht - 2021
Veranstaltung35th Conference on Neural Information Processing Systems, NeurIPS 2021 - Virtual, Online
Dauer: 6 Dez. 202114 Dez. 2021

Publikationsreihe

NameAdvances in Neural Information Processing Systems
Band9
ISSN (Print)1049-5258

Konferenz

Konferenz35th Conference on Neural Information Processing Systems, NeurIPS 2021
OrtVirtual, Online
Zeitraum6/12/2114/12/21

Fingerprint

Untersuchen Sie die Forschungsthemen von „GemNet: Universal Directional Graph Neural Networks for Molecules“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren