ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations
This work addresses the need for efficient and accurate machine learning models in quantum physics simulations, offering a novel approach that could impact large-scale computational chemistry and materials science.
The authors tackled the problem of approximating atomic forces in quantum calculations by developing ForceNet, a graph neural network that avoids explicit physical constraints, achieving higher accuracy and faster training/inference than state-of-the-art physics-based GNNs on the large OC20 dataset.
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.