Spatial Attention Kinetic Networks with E(n)-Equivariance

Abstract

Neural networks that are equivariant to rotations, translations, reflections, and permutations on n-dimensional geometric space have shown promise in physical modeling for tasks such as accurately but inexpensively modeling complex potential energy surfaces to guiding the sampling of complex dynamical systems or forecasting their time evolution. Current state-of-the-art methods employ spherical harmonics to encode higher-order interactions among particles, which are computationally expensive. In this paper, we propose a simple alternative functional form that uses neurally parametrized linear combinations of edge vectors to achieve equivariance while still universally approximating node environments. Incorporating this insight, we design spatial attention kinetic networks with E(n)-equivariance, or SAKE, which are competitive in many-body system modeling tasks while being significantly faster.

Cited by in corpus (4)

• On the design space between molecular mechanics and machine learning force fields
• Scaling Graph Neural Networks to Large Proteins
• Molecular Dynamics and Machine Learning Unlock Possibilities in Beauty Design -- A Perspective
• Knowledge Distillation of a Protein Language Model Yields a Foundational Implicit Solvent Model