Geometry-enhanced Pre-training on Interatomic Potentials

TL;DR

This paper introduces a geometry-enhanced pre-training approach for machine learning interatomic potentials that improves accuracy and transferability in molecular dynamics simulations by leveraging unlabeled 3D configurations and geometric learning.

Contribution

The authors propose a novel two-stage geometric structure learning paradigm with self-supervised techniques to enhance MLIPs, especially for complex molecular systems, with minimal additional computational costs.

Findings

Significant accuracy improvements on various molecular benchmarks.

Enhanced generalization of MLIPs across different architectures.

Effective use of unlabeled 3D configurations for pre-training.

Abstract

Machine learning interatomic potentials (MLIPs) enables molecular dynamics (MD) simulations with ab initio accuracy and has been applied to various fields of physical science. However, the performance and transferability of MLIPs are limited by insufficient labeled training data, which require expensive ab initio calculations to obtain the labels, especially for complex molecular systems. To address this challenge, we design a novel geometric structure learning paradigm that consists of two stages. We first generate a large quantity of 3D configurations of target molecular system with classical molecular dynamics simulations. Then, we propose geometry-enhanced self-supervised learning consisting of masking, denoising, and contrastive learning to better capture the topology and 3D geometric information from the unlabeled 3D configurations. We evaluate our method on various benchmarks ranging from small molecule datasets to complex periodic molecular systems with more types of elements. The experimental results show that the proposed pre-training method can greatly enhance the accuracy of MLIPs with few extra computational costs and works well with different invariant or equivariant graph neural network architectures. Our method improves the generalization capability of MLIPs and helps to realize accurate MD simulations for complex molecular systems.