High-performance training and inference for deep equivariant interatomic potentials
This work addresses the growing demands for scalable software in atomistic modeling, benefiting researchers in computational chemistry and materials science, though it is incremental as it builds on existing frameworks.
The authors tackled the challenge of scaling deep equivariant neural networks for interatomic potentials by overhauling the NequIP framework to improve multi-node parallelism and computational performance, resulting in up to an 18x speedup in molecular dynamics calculations on relevant system sizes.
Machine learning interatomic potentials, particularly those based on deep equivariant neural networks, have demonstrated state-of-the-art accuracy and computational efficiency in atomistic modeling tasks like molecular dynamics and high-throughput screening. The size of datasets and demands of downstream workflows are growing rapidly, making robust and scalable software essential. This work presents a major overhaul of the NequIP framework focusing on multi-node parallelism, computational performance, and extensibility. The redesigned framework supports distributed training on large datasets and removes barriers preventing full utilization of the PyTorch 2.0 compiler at train time. We demonstrate this acceleration in a case study by training Allegro models on the SPICE 2 dataset of organic molecular systems. For inference, we introduce the first end-to-end infrastructure that uses the PyTorch Ahead-of-Time Inductor compiler for machine learning interatomic potentials. Additionally, we implement a custom kernel for the Allegro model's most expensive operation, the tensor product. Together, these advancements speed up molecular dynamics calculations on system sizes of practical relevance by up to a factor of 18.