Ewald-Based Long-Range Message Passing for Molecular Graphs | Arthur Kosmala
Try datamol.io - the open source toolkit that simplifies molecular processing and featurization workflows for machine learning scientists working in drug discovery: https://datamol.io/
Never miss another LoGG Talk, add the schedule to your calendar: https://m2d2.io/talks/m2d2/about/
Join the Learning on Graphs and Geometry Reading Group on Slack: https://join.slack.com/t/logag/shared...
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 cheap and agnostic to other architectural details. We test the approach with four baseline models and two datasets containing diverse periodic (OC20) and aperiodic structures (OE62). We observe robust improvements in energy mean absolute errors across all models and datasets, 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.
Speaker: Arthur Kosmala - / arthur-kosmala-9219371b2
Twitter Hannes: / hannesstaerk
Twitter Dominique: / dom_beaini
Twitter datamol.io: / datamol_io
~
Chapters
00:00 - Intro
02:42 - Problem Setting
05:05 - MPNN Blueprint
07:21 - Continous-Filter Convolutions
11:30 - Periodic Boundary Conditions
26:15 - Ewald Summary
35:29 - Ewald Message Passing
38:23 - Long-Range Message Sums
49:54 - The Aperiodic Case
53:22 - Radial vs. Non-Radial Filtering
57:21 - Combination with Existing Models
59:08 - Baselines & Datasets
1:01:47 - Comparison Studies
1:06:48 - Analysis of Long-Range Impact
1:09:13 - Force vs. Energy Results
1:10:41 - Conclusion
1:12:33 - Q+A