Learning local equivariant representations for large-scale atomistic dynamics

By Albert Musaelian, Simon L. Batzner, Anders Johansson, Lixin Sun, Cameron J. Owen, Mordechai Kornbluth, Boris Kozinsky. Nature Communications, 2022

820 citations recorded by the source corpus.

A simultaneously accurate and computationally efficient parametrization of the potential energy surface of molecules and materials is a long-standing goal in the natural sciences. While atom-centered message passing neural networks (MPNNs) have shown remarkable accuracy, their information propagation has limited the accessible length-scales. Local methods, conversely, scale to large simulations but have suffered from inferior accuracy. This work introduces Allegro, a strictly local equivariant deep neural network interatomic potential architecture that simultaneously exhibits excellent accuracy and scalability. Allegro represents a many-body potential using iterated tensor products of learned equivariant representations without atom-centered message passing. Allegro obtains improvements over state-of-the-art methods on QM9 and revMD17. A single tensor product layer outperforms existing deep MPNNs and transformers on QM9. Furthermore, Allegro displays remarkable generalization to out-of-distribution data. Molecular simulations using Allegro recover structural and kinetic properties of an amorphous electrolyte in excellent agreement with ab-initio simulations. Finally, we demonstrate parallelization with a simulation of 100 million atoms. The paper presents a method that allows scaling machine learning interatomic potentials to extremely large systems, while at the same time retaining the remarkable accuracy and learning efficiency of deep equivariant models. This is obtained with an E(3)- equivariant neural network architecture that combines the high accuracy of equivariant neural networks with the scalability of local methods.

Subfields: Artificial Intelligence, Materials Chemistry, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/aeb87738cf0058e1c85cad38d5cda241d420a455

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Materials Chemistry

In Molecular Biology

Citations in the corpus

MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields (2022, methodology): “A severe limitation of many previously published MPNNs was their large receptive ﬁeld, making it difﬁcult to parallelize the evaluation across multiple GPUs.”
Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs (2022, background): “Equivariant neural networks [72, 43, 80, 23, 50, 74, 4, 38, 65, 63, 77, 5, 71, 46, 51] operate on geometric tensors like type-L vectors to achieve equivariance.”
e3nn: Euclidean Neural Networks (2022, neutral): “…from micro-CT images [5], approximating steady-state ﬂuid ﬂow around a ﬁxed particle [27], deﬁning a graph neural network for which node and edge attributes are not restricted to invariant scalars [4], generate coarse-graining of molecular conformations [32] and simulate 100 million atoms [20].”
The design space of E(3)-equivariant atom-centred interatomic potentials (2022, neutral): “We denote the set of neighbors of an atom i , i.e., atoms with the cutoﬀ distance, by N ( i ).”