Neural P$^3$M: A Long-Range Interaction Modeling Enhancer for Geometric GNNs
This addresses a bottleneck in molecular modeling for computational chemistry, though it appears incremental as an enhancer to existing methods.
The paper tackles the problem of geometric GNNs struggling with long-range interactions in large molecules by introducing Neural P^3M, which enhances these models by adding mesh points and reimagining operations, resulting in improved accuracy on benchmarks like MD22 and a 22% average gain on OE62.
Geometric graph neural networks (GNNs) have emerged as powerful tools for modeling molecular geometry. However, they encounter limitations in effectively capturing long-range interactions in large molecular systems. To address this challenge, we introduce Neural P$^3$M, a versatile enhancer of geometric GNNs to expand the scope of their capabilities by incorporating mesh points alongside atoms and reimaging traditional mathematical operations in a trainable manner. Neural P$^3$M exhibits flexibility across a wide range of molecular systems and demonstrates remarkable accuracy in predicting energies and forces, outperforming on benchmarks such as the MD22 dataset. It also achieves an average improvement of 22% on the OE62 dataset while integrating with various architectures.