ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations

TL;DR

ForceNet is a scalable graph neural network that predicts atomic forces more accurately and efficiently than existing physics-based models by implicitly enforcing physical constraints through data augmentation.

Contribution

We introduce ForceNet, a novel GNN that forgoes explicit physical constraints, achieving better accuracy and efficiency on large-scale quantum physics data.

Findings

ForceNet outperforms state-of-the-art GNNs in force prediction accuracy.

ForceNet is faster in training and inference compared to existing models.

Implicit physical constraints via data augmentation are effective.

Abstract

With massive amounts of atomic simulation data available, there is a huge opportunity to develop fast and accurate machine learning models to approximate expensive physics-based calculations. The key quantity to estimate is atomic forces, where the state-of-the-art Graph Neural Networks (GNNs) explicitly enforce basic physical constraints such as rotation-covariance. However, to strictly satisfy the physical constraints, existing models have to make tradeoffs between computational efficiency and model expressiveness. Here we explore an alternative approach. By not imposing explicit physical constraints, we can flexibly design expressive models while maintaining their computational efficiency. Physical constraints are implicitly imposed by training the models using physics-based data augmentation. To evaluate the approach, we carefully design a scalable and expressive GNN model, ForceNet, and apply it to OC20 (Chanussot et al., 2020), an unprecedentedly-large dataset of quantum physics calculations. Our proposed ForceNet is able to predict atomic forces more accurately than state-of-the-art physics-based GNNs while being faster both in training and inference. Overall, our promising and counter-intuitive results open up an exciting avenue for future research.