Completeness of Atomic Structure Representations

Abstract

In this paper, we address the challenge of obtaining a comprehensive and symmetric representation of point particle groups, such as atoms in a molecule, which is crucial in physics and theoretical chemistry. The problem has become even more important with the widespread adoption of machine-learning techniques in science, as it underpins the capacity of models to accurately reproduce physical relationships while being consistent with fundamental symmetries and conservation laws. However, some of the descriptors that are commonly used to represent point clouds -- most notably those based on discretized correlations of the neighbor density, that underpin most of the existing ML models of matter at the atomic scale -- are unable to distinguish between special arrangements of particles in three dimensions. This makes it impossible to machine learn their properties. Atom-density correlations are provably complete in the limit in which they simultaneously describe the mutual relationship between all atoms, which is impractical. We present a novel approach to construct descriptors of \emph{finite} correlations based on the relative arrangement of particle triplets, which can be employed to create symmetry-adapted models with universal approximation capabilities, which have the resolution of the neighbor discretization as the sole convergence parameter. Our strategy is demonstrated on a class of atomic arrangements that are specifically built to defy a broad class of conventional symmetric descriptors, showcasing its potential for addressing their limitations.

References in corpus (34)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Machine learning and the physical sciences
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Recent developments in the PySCF program package
• Euclidean Distance Matrices: Essential Theory, Algorithms and Applications
• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
• Physics-inspired structural representations for molecules and materials
• On the Completeness of Atomic Structure Representations
• Atom-Density Representations for Machine Learning
• The energy landscape of fullerene materials: a comparison between boron, boron-nitride and carbon
• Efficient nonparametric $n$-body force fields from machine learning
• Metrics for measuring distances in configuration spaces
• Cormorant: Covariant Molecular Neural Networks
• Transferable atomic multipole machine learning models for small organic molecules
• A fingerprint based metric for measuring similarities of crystalline structures
• The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentials
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Equivariant representations for molecular Hamiltonians and N-center atomic-scale properties
• Comment on "Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning"
• An assessment of the structural resolution of various fingerprints commonly used in machine learning
• Scalars are universal: Equivariant machine learning, structured like classical physics
• Manifolds of quasi-constant SOAP and ACSF fingerprints and the resulting failure to machine learn four-body interactions
• Smooth, exact rotational symmetrization for deep learning on point clouds
• Frame Averaging for Invariant and Equivariant Network Design
• Polynomial-time algorithms for continuous metrics on atomic clouds of unordered points
• Comment on "Manifolds of quasi-constant SOAP and ACSF fingerprints and the resulting failure to machine learn four body interactions"
• A smooth basis for atomistic machine learning
• Exactly computable and continuous metrics on isometry classes of finite and 1-periodic sequences
• Recognizing Rigid Patterns of Unlabeled Point Clouds by Complete and Continuous Isometry Invariants with no False Negatives and no False Positives

Cited by in corpus (5)

• Machine Learning in QM/MM Molecular Dynamics Simulations of Condensed-Phase Systems
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Symmetry-invariant quantum machine learning force fields
• Representing spherical tensors with scalar-based machine-learning models
• Exact average many-body interatomic interaction model for random alloys