A Universal Framework for Featurization of Atomistic Systems

by   Xiangyun Lei, et al.

Molecular dynamics simulations are an invaluable tool in numerous scientific fields. However, the ubiquitous classical force fields cannot describe reactive systems, and quantum molecular dynamics are too computationally demanding to treat large systems or long timescales. Reactive force fields based on physics or machine learning can be used to bridge the gap in time and length scales, but these force fields require substantial effort to construct and are highly specific to given chemical composition and application. The extreme flexibility of machine learning models promises to yield reactive force fields that provide a more general description of chemical bonding. However, a significant limitation of machine learning models is the use of element-specific features, leading to models that scale poorly with the number of elements. This work introduces the Gaussian multi-pole (GMP) featurization scheme that utilizes physically-relevant multi-pole expansions of the electron density around atoms to yield feature vectors that interpolate between element types and have a fixed dimension regardless of the number of elements present. We combine GMP with neural networks to directly compare it to the widely-used Behler-Parinello symmetry functions for the MD17 dataset, revealing that it exhibits improved accuracy and computational efficiency. Further, we demonstrate that GMP-based models can achieve chemical accuracy for the QM9 dataset, and their accuracy remains reasonable even when extrapolating to new elements. Finally, we test GMP-based models for the Open Catalysis Project (OCP) dataset, revealing comparable performance and improved learning rates when compared to graph convolutional deep learning models. The results indicate that this featurization scheme fills a critical gap in the construction of efficient and transferable reactive force fields.


Symmetry-adapted graph neural networks for constructing molecular dynamics force fields

Molecular dynamics is a powerful simulation tool to explore material pro...

By-passing the Kohn-Sham equations with machine learning

Last year, at least 30,000 scientific papers used the Kohn-Sham scheme o...

BLAST: Bridging Length/time scales via Atomistic Simulation Toolkit

The ever-increasing power of supercomputers coupled with highly scalable...

Espaloma-0.3.0: Machine-learned molecular mechanics force field for the simulation of protein-ligand systems and beyond

Molecular mechanics (MM) force fields – the models that characterize the...

SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effects

In recent years, machine-learned force fields (ML-FFs) have gained incre...

Machine Learning Force Fields

In recent years, the use of Machine Learning (ML) in computational chemi...

Fast Bayesian Force Fields from Active Learning: Study of Inter-Dimensional Transformation of Stanene

We present a way to dramatically accelerate Gaussian process models for ...

Please sign up or login with your details

Forgot password? Click here to reset