Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements

Abstract

Machine-learning of atomic-scale properties amounts to extracting correlations between structure, composition and the quantity that one wants to predict. Representing the input structure in a way that best reflects such correlations makes it possible to improve the accuracy of the model for a given amount of reference data. When using a description of the structures that is transparent and well-principled, optimizing the representation might reveal insights into the chemistry of the data set. Here we show how one can generalize the SOAP kernel to introduce a distance-dependent weight that accounts for the multi-scale nature of the interactions, and a description of correlations between chemical species. We show that this improves substantially the performance of ML models of molecular and materials stability, while making it easier to work with complex, multi-component systems and to extend SOAP to coarse-grained intermolecular potentials. The element correlations that give the best performing model show striking similarities with the conventional periodic table of the elements, providing an inspiring example of how machine learning can rediscover, and generalize, intuitive concepts that constitute the foundations of chemistry.

arXiv comment9 pages, 4 figures

References in corpus (24)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• SchNet - a deep learning architecture for molecules and materials
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Big Data meets Quantum Chemistry Approximations: The $\Delta$-Machine Learning Approach
• Comparing molecules and solids across structural and alchemical space
• Machine Learning Unifies the Modelling of Materials and Molecules
• Machine learning prediction errors better than DFT accuracy
• Machine-learning based interatomic potential for amorphous carbon
• Machine Learning Energies of 2 M Elpasolite (ABC${}_2$D${}_6$) Crystals
• Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
• Alchemical and structural distribution based representation for improved QML
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• Hierarchical modeling of molecular energies using a deep neural network
• WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
• Achieving DFT accuracy with a machine-learning interatomic potential: thermomechanics and defects in bcc ferromagnetic iron
• Understanding molecular representations in machine learning: The role of uniqueness and target similarity
• Accurate Interatomic Force Fields via Machine Learning with Covariant Kernels
• Atom2Vec: learning atoms for materials discovery
• Comparison of permutationally invariant polynomials, neural networks, and Gaussian approximation potentials in representing water interactions through many-body expansions
• Efficient nonparametric $n$-body force fields from machine learning

Cited by in corpus (45)

• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• On-the-fly machine learning force field generation: Application to melting points
• Physics-inspired structural representations for molecules and materials
• An Accurate and Transferable Machine Learning Potential for Carbon
• On the Completeness of Atomic Structure Representations
• Evaluation of the MACE Force Field Architecture: from Medicinal Chemistry to Materials Science
• Unsupervised machine learning in atomistic simulations, between predictions and understanding
• Atom-Density Representations for Machine Learning
• Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Learning the electronic density of states in condensed matter
• Predicting molecular dipole moments by combining atomic partial charges and atomic dipoles
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Simple machine-learned interatomic potentials for complex alloys
• Equation of state of fluid methane from first principles with machine learning potentials
• A Bayesian approach to NMR crystal structure determination
• Equivariant representations for molecular Hamiltonians and N-center atomic-scale properties
• Equivariant analytical mapping of first principles Hamiltonians to accurate and transferable materials models
• Efficient implementation of atom-density representations
• Developments and Further Applications of Ephemeral Data Derived Potentials
• Combining phonon accuracy with high transferability in Gaussian approximation potential models
• A New Kind of Atlas of Zeolite Building Blocks
• Tensor-reduced atomic density representations
• Neural Network Kinetics for Exploring Diffusion Multiplicity and Chemical Ordering in Compositionally Complex Materials
• Predicting the Charge Density Response in Metal Electrodes
• Multilayer atomic cluster expansion for semi-local interactions
• Phonon-induced localization of excitons in molecular crystals from first principles
• Kernel based quantum machine learning at record rate : Many-body distribution functionals as compact representations
• Optimal radial basis for density-based atomic representations
• Resolution limit of data-driven coarse-grained models spanning chemical space
• Atomic structure optimization with machine-learning enabled interpolation between chemical elements
• Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)
• Predicting hot-electron free energies from ground-state data
• Metatensor and metatomic: foundational libraries for interoperable atomistic machine learning
• Machine learning at the atomic-scale
• A smooth basis for atomistic machine learning
• Comment on "Manifolds of quasi-constant SOAP and ACSF fingerprints and the resulting failure to machine learn four body interactions"
• Compressing physical properties of atomic species for improving predictive chemistry
• Hybrid localized graph kernel for machine learning energy-related properties of molecules and solids
• Structure-Driven Prediction of Magnetic Order in Uranium Compounds
• Resolving the Body-Order Paradox of Machine Learning Interatomic Potentials
• Representing spherical tensors with scalar-based machine-learning models
• Efficient interpolation of molecular properties across chemical compound space with low-dimensional descriptors
• Compressing local atomic neighbourhood descriptors