Efficient Machine Learning Interatomic Potentials
Organizations: ICAMS, Ruhr-Universität Bochum, Bochum, Germany
Abstract
Machine learning interatomic potentials (MLIPs) reach an accuracy close to quantum mechanical reference calculations at a small fraction of the cost. Efficiency is the purpose of MLIPs, yet strategies for efficient MLIPs have not been reviewed systematically. This review takes efficiency as its guideline. To compare MLIPs on a common footing, we represent them in a complete basis expansion, the Atomic Cluster Expansion (ACE) and its graph generalization (GRACE). The expansion is canonical, built from products of single-particle functions. Classical interatomic potentials, such as the embedded-atom method, and many modern MLIPs, including equivariant transformers, can be represented within this expansion. Much of the apparent diversity of MLIP architectures therefore reflects how non-linearity, layer coupling and optimization are handled rather than which interactions the models can represent. We show how the cost of an MLIP becomes linear in system size and, at fixed rank of the tensor decomposition, in model complexity, and we discuss the role of machine learning, which provides non-linear parameterizations and optimization. Because efficiency is meaningful only at a given accuracy, we also compare accuracy and cost. On the Matbench Discovery leaderboard, models with three to five layers approach the accuracy of deeper models, and one basis, truncated along the Pareto front of error and speed, spans more than two orders of magnitude in cost. Ultimately, training data determine the quality of any MLIP.
Figures & tables
| Property | Naive | Efficient |
|---|---|---|
| Atoms | ||
| Neighbors | ||
| Basis functions | ||
| Body order | ||
| Graph layers | ||
| Chemical species |
| Model | Symmetry | Non-linearity | Many-body | Maps onto | |||
| Classical potentials | |||||||
| EAM (Cu) / FS / EMT | invariant | embedding | expl. | 1 | 0 | 2-body ACE embedding | |
| Tersoff (Si) / MEAM | invariant | bond order / embedding | expl. | 2 | / 3 | / screened | 3-body ACE non-linear function |
| Local descriptors and MLIPs | |||||||
| ACE (PACE, Cu) | invariant | linear or FS-like | expl. | 6 | 6 | local basis, truncated | |
| HDNNP (Si) / ANI (1x) | invariant | neural network | learned | 2 | / | 2-/3-body ACE NN | |
Machine Learning Interatomic Potentials
Machine learning interatomic potentials predict atomic energies and forces for fast simulations of molecules and materials.
See all 107 papers