Pre-training Graph Neural Networks with Structural Fingerprints for Materials Discovery
This addresses the scalability problem for researchers in materials science by enabling more efficient pre-training of GNNs for large-scale atomistic data.
The paper tackles the computational cost of pre-training graph neural networks for materials science by proposing a novel objective that uses cheaply-computed structural fingerprints as targets, achieving comparable performance to methods requiring expensive quantum mechanical calculations.
In recent years, pre-trained graph neural networks (GNNs) have been developed as general models which can be effectively fine-tuned for various potential downstream tasks in materials science, and have shown significant improvements in accuracy and data efficiency. The most widely used pre-training methods currently involve either supervised training to fit a general force field or self-supervised training by denoising atomic structures equilibrium. Both methods require datasets generated from quantum mechanical calculations, which quickly become intractable when scaling to larger datasets. Here we propose a novel pre-training objective which instead uses cheaply-computed structural fingerprints as targets while maintaining comparable performance across a range of different structural descriptors. Our experiments show this approach can act as a general strategy for pre-training GNNs with application towards large scale foundational models for atomistic data.