Problem
Research questions and friction points this paper is trying to address.
multiscale interactions
machine learning force fields
long-range many-body effects
coarse-graining
quantum-mechanical interactions
Machine Learning Multiscale Interactions
🤖 AI Summary
Current machine learning force fields struggle to accurately model long-range many-body interactions across multiple scales, limiting their predictive capability for complex systems such as biomolecules and nanostructures. This work proposes MuSE, a novel model that introduces a soft coarse-graining pooling mechanism, which constructs hierarchical representations through smooth fractional assignments from atoms to coarse-grained nodes, enabling architecture-agnostic multiscale modeling. The approach seamlessly integrates with mainstream force fields—including SO3krates, MACE, and PaiNN—to consistently capture both short- and long-range quantum mechanical effects. Evaluated on benchmarks including Hessian spectra, protein folding trajectories, and molecule–graphene energy profiles, MuSE significantly outperforms existing long-range machine learning force fields, accurately resolving multiscale physical features.
📝 Abstract
Realistic physical systems are characterised by emergent interactions across multiple length and time scales, posing a significant challenge for predictive machine learning (ML) models. Most scientific ML models focus on a narrow range of interactions. While machine learning force fields (MLFFs) offer near-quantum accuracy, the ubiquitous message-passing layers miss long-range many-body effects. Here we introduce the Multiscale Structural Ensemble (MuSE), a hierarchical model that uses Soft Coarse-Graining Pooling to construct coarse representations from smooth fractional assignments of atoms to coarse nodes, enabling MLFF modules to operate across multiple scales. MuSE is architecture-agnostic and coupled with SO3krates, MACE, and PaiNN MLFFs for both molecules and materials. We demonstrate the power of MuSE through Hessian-based benchmarks, folding trajectories for biomolecules, and energy profiles in molecule-graphene nanostructures, where MuSE accurately captures quantum-mechanical interactions at relevant scales -- unlike other recent long-range ML models.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Multiscale Modeling
Machine Learning Force Fields
Soft Coarse-Graining
Long-range Many-body Interactions
Hierarchical Representation
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
