Physics-inspired Equivariant Descriptors of Non-bonded Interactions
This addresses a bottleneck in materials and molecular modeling for researchers, offering a coherent approach to incorporate non-bonded interactions, though it appears incremental as an extension of existing methods.
The authors tackled the neglect of long-range interactions like electrostatics in machine learning for atomistic modeling by extending the LODE framework to handle diverse long-range effects, providing a physical interpretation via multipole expansion and demonstrating it on toy systems and molecular dimers.
One essential ingredient in many machine learning (ML) based methods for atomistic modeling of materials and molecules is the use of locality. While allowing better system-size scaling, this systematically neglects long-range (LR) effects, such as electrostatics or dispersion interaction. We present an extension of the long distance equivariant (LODE) framework that can handle diverse LR interactions in a consistent way, and seamlessly integrates with preexisting methods by building new sets of atom centered features. We provide a direct physical interpretation of these using the multipole expansion, which allows for simpler and more efficient implementations. The framework is applied to simple toy systems as proof of concept, and a heterogeneous set of molecular dimers to push the method to its limits. By generalizing LODE to arbitrary asymptotic behaviors, we provide a coherent approach to treat arbitrary two- and many-body non-bonded interactions in the data-driven modeling of matter.