Ewald-based Long-Range Message Passing for Molecular Graphs

Arthur Kosmala, Johannes Gasteiger, Nicholas Gao, Stephan Günnemann
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:17544-17563, 2023.

Abstract

Neural architectures that learn potential energy surfaces from molecular data have undergone fast improvement in recent years. A key driver of this success is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling with system size partly relies upon a spatial distance limit on messages. While this focus on locality is a useful inductive bias, it also impedes the learning of long-range interactions such as electrostatics and van der Waals forces. To address this drawback, we propose Ewald message passing: a nonlocal Fourier space scheme which limits interactions via a cutoff on frequency instead of distance, and is theoretically well-founded in the Ewald summation method. It can serve as an augmentation on top of existing MPNN architectures as it is computationally inexpensive and agnostic to architectural details. We test the approach with four baseline models and two datasets containing diverse periodic (OC20) and aperiodic structures (OE62). Across all models and datasets, we observe robust improvements in energy mean absolute errors, averaging 10% on OC20 and 16% on OE62. Our analysis shows an outsize impact of these improvements on structures with high long-range contributions to the ground-truth energy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-kosmala23a, title = {Ewald-based Long-Range Message Passing for Molecular Graphs}, author = {Kosmala, Arthur and Gasteiger, Johannes and Gao, Nicholas and G\"{u}nnemann, Stephan}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {17544--17563}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/kosmala23a/kosmala23a.pdf}, url = {https://proceedings.mlr.press/v202/kosmala23a.html}, abstract = {Neural architectures that learn potential energy surfaces from molecular data have undergone fast improvement in recent years. A key driver of this success is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling with system size partly relies upon a spatial distance limit on messages. While this focus on locality is a useful inductive bias, it also impedes the learning of long-range interactions such as electrostatics and van der Waals forces. To address this drawback, we propose Ewald message passing: a nonlocal Fourier space scheme which limits interactions via a cutoff on frequency instead of distance, and is theoretically well-founded in the Ewald summation method. It can serve as an augmentation on top of existing MPNN architectures as it is computationally inexpensive and agnostic to architectural details. We test the approach with four baseline models and two datasets containing diverse periodic (OC20) and aperiodic structures (OE62). Across all models and datasets, we observe robust improvements in energy mean absolute errors, averaging 10% on OC20 and 16% on OE62. Our analysis shows an outsize impact of these improvements on structures with high long-range contributions to the ground-truth energy.} }
Endnote
%0 Conference Paper %T Ewald-based Long-Range Message Passing for Molecular Graphs %A Arthur Kosmala %A Johannes Gasteiger %A Nicholas Gao %A Stephan Günnemann %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-kosmala23a %I PMLR %P 17544--17563 %U https://proceedings.mlr.press/v202/kosmala23a.html %V 202 %X Neural architectures that learn potential energy surfaces from molecular data have undergone fast improvement in recent years. A key driver of this success is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling with system size partly relies upon a spatial distance limit on messages. While this focus on locality is a useful inductive bias, it also impedes the learning of long-range interactions such as electrostatics and van der Waals forces. To address this drawback, we propose Ewald message passing: a nonlocal Fourier space scheme which limits interactions via a cutoff on frequency instead of distance, and is theoretically well-founded in the Ewald summation method. It can serve as an augmentation on top of existing MPNN architectures as it is computationally inexpensive and agnostic to architectural details. We test the approach with four baseline models and two datasets containing diverse periodic (OC20) and aperiodic structures (OE62). Across all models and datasets, we observe robust improvements in energy mean absolute errors, averaging 10% on OC20 and 16% on OE62. Our analysis shows an outsize impact of these improvements on structures with high long-range contributions to the ground-truth energy.
APA
Kosmala, A., Gasteiger, J., Gao, N. & Günnemann, S.. (2023). Ewald-based Long-Range Message Passing for Molecular Graphs. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:17544-17563 Available from https://proceedings.mlr.press/v202/kosmala23a.html.

Related Material