A foundation model for atomistic materials chemistry

Ilyes Batatia, Philipp Benner, Yuan Chiang, …, Lars L. Schaaf, …, Gábor Csányi

arXiv preprint · 88 authors, 2024

arXiv Models

Key contributions

• One model, not one model per system. Early ML force fields needed a bespoke potential built and validated for each system of interest, with little transfer between them. MACE-MP-0 is a single general-purpose model trained on a public dataset of moderate size.
• Stable MD across the physical sciences. Demonstrated on solids, liquids, gases, chemical reactions, interfaces, and the dynamics of a small protein — qualitatively, and at times quantitatively, accurate.
• A foundation model you can fine-tune. Usable out of the box, and when ab initio accuracy is required it can be fine-tuned on just a handful of application-specific data points.
• A lower barrier to entry. Experienced users get reliable results much faster; beginners no longer have to build a potential before they can simulate anything. The video above is the whole workflow: pip install mace-torch, load the model, run finite-temperature MD.

Abstract

Atomistic simulations of matter, especially those that leverage first-principles (ab initio) electronic structure theory, provide a microscopic view of the world, underpinning much of our understanding of chemistry and materials science. Over the last decade or so, machine-learned force fields have transformed atomistic modeling by enabling simulations of ab initio quality over unprecedented time and length scales. However, early ML force fields have largely been limited by: (i) the substantial computational and human effort of developing and validating potentials for each particular system of interest; and (ii) a general lack of transferability from one chemical system to the next.

Here we show that it is possible to create a general-purpose atomistic ML model, trained on a public dataset of moderate size, that is capable of running stable molecular dynamics for a wide range of molecules and materials. We demonstrate the power of the MACE-MP-0 model — and its qualitative and at times quantitative accuracy — on a diverse set of problems in the physical sciences, including properties of solids, liquids, gases, chemical reactions, interfaces and even the dynamics of a small protein. The model can be applied out of the box as a starting or “foundation” model for any atomistic system of interest and, when desired, can be fine-tuned on just a handful of application-specific data points to reach ab initio accuracy. Establishing that a stable force-field model can cover almost all materials changes atomistic modeling in a fundamental way: experienced users get reliable results much faster, and beginners face a lower barrier to entry. Foundation models thus represent a step towards democratising the revolution in atomic-scale modeling that has been brought about by ML force fields.

Citation

@article{batatia2024foundation,
  title   = {A foundation model for atomistic materials chemistry},
  author  = {Batatia, Ilyes and Benner, Philipp and Chiang, Yuan and others},
  journal = {arXiv preprint arXiv:2401.00096},
  year    = {2024}
}