Comparing molecules and solids across structural and alchemical space

Abstract

Evaluating the (dis)similarity of crystalline, disordered and molecular compounds is a critical step in the development of algorithms to navigate automatically the configuration space of complex materials. For instance, a structural similarity metric is crucial for classifying structures, searching chemical space for better compounds and materials, and driving the next generation of machine-learning techniques for predicting the stability and properties of molecules and materials. In the last few years several strategies have been designed to compare atomic coordination environments. In particular, the Smooth Overlap of Atomic Positions (SOAP) has emerged as an elegant framework to obtain translation, rotation and permutation-invariant descriptors of groups of atoms, driven by the design of various classes of machine-learned inter-atomic potentials. Here we discuss how one can combine such local descriptors using a Regularized Entropy Match (REMatch) approach to describe the similarity of both whole molecular and bulk periodic structures, introducing powerful metrics that enable the navigation of alchemical and structural complexity within a unified framework. Furthermore, using this kernel and a ridge regression method we can predict atomization energies for a database of small organic molecules with a mean absolute error below 1kcal/mol, reaching an important milestone in the application of machine-learning techniques to the evaluation of molecular properties.

18 pages, 7 figures

References in corpus (9)

• Big Data of Materials Science - Critical Role of the Descriptor
• AiiDA: Automated Interactive Infrastructure and Database for Computational Science
• Interatomic potentials for ionic systems with density functional accuracy based on charge densities obtained by a neural network
• Materials Cartography: Representing and Mining Material Space Using Structural and Electronic Fingerprints
• A learning scheme to predict atomic forces and accelerate materials simulations
• Accelerated materials property predictions and design using motif-based fingerprints
• Machine learning for many-body physics: The case of the Anderson impurity model
• A fingerprint based metric for measuring similarities of crystalline structures
• Low-density silicon allotropes for photovoltaic applications

Cited by in corpus (198)

• Machine Learning Force Fields
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• A Performance and Cost Assessment of Machine Learning Interatomic Potentials
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• Machine Learning Unifies the Modelling of Materials and Molecules
• Data-driven materials science: status, challenges and perspectives
• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• Machine learning prediction errors better than DFT accuracy
• Machine-learning based interatomic potential for amorphous carbon
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
• Physics-inspired structural representations for molecules and materials
• Machine learning a general purpose interatomic potential for silicon
• SchNetPack: A Deep Learning Toolbox For Atomistic Systems
• Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
• FCHL revisited: faster and more accurate quantum machine learning
• Alchemical and structural distribution based representation for improved QML
• Resolving transition metal chemical space: feature selection for machine learning and structure-property relationships
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• Hierarchical modeling of molecular energies using a deep neural network
• Automatic Selection of Atomic Fingerprints and Reference Configurations for Machine-Learning Potentials
• Unified Representation of Molecules and Crystals for Machine Learning
• Achieving DFT accuracy with a machine-learning interatomic potential: thermomechanics and defects in bcc ferromagnetic iron
• Chemical Shifts in Molecular Solids by Machine Learning
• An Accurate and Transferable Machine Learning Potential for Carbon
• A Transferable Machine-Learning Model of the Electron Density
• Understanding molecular representations in machine learning: The role of uniqueness and target similarity
• sGDML: Constructing Accurate and Data Efficient Molecular Force Fields Using Machine Learning
• Symmetries and many-body excited states with neural-network quantum states
• De novo exploration and self-guided learning of potential-energy surfaces
• Incorporating long-range physics in atomic-scale machine learning
• Extending the Accuracy of the SNAP Interatomic Potential Form
• A Machine Learning Potential for Graphene
• Accurate molecular polarizabilities with coupled-cluster theory and machine learning
• Perspective on integrating machine learning into computational chemistry and materials science
• Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks
• Unsupervised machine learning in atomistic simulations, between predictions and understanding
• Atom-Density Representations for Machine Learning
• Comparison of permutationally invariant polynomials, neural networks, and Gaussian approximation potentials in representing water interactions through many-body expansions
• Fast and Accurate Uncertainty Estimation in Chemical Machine Learning
• Machine learning of molecular properties: locality and active learning
• Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials
• Constrained crystals deep convolutional generative adversarial network for the inverse design of crystal structures
• Constructing first-principles phase diagrams of amorphous LixSi using machine-learning-assisted sampling with an evolutionary algorithm
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Efficient nonparametric $n$-body force fields from machine learning
• Learning the electronic density of states in condensed matter
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements
• QM7-X: A comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules
• Predicting polymorphism in molecular crystals using orientational entropy
• Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces
• Performance Assessment of Universal Machine Learning Interatomic Potentials: Challenges and Directions for Materials' Surfaces
• Training neural nets to learn reactive potential energy surfaces using interactive quantum chemistry in virtual reality
• Updates to the DScribe Library: New Descriptors and Derivatives
• Transfer learning for materials informatics using crystal graph convolutional neural network
• Gaussian approximation potential modeling of lithium intercalation in carbon nanostructures
• Error-Controlled Exploration of Chemical Reaction Networks with Gaussian Processes
• Solid Harmonic Wavelet Scattering for Predictions of Molecule Properties
• Graph Neural Network Predictions of Metal Organic Framework CO2 Adsorption Properties
• Sparse Gaussian Process Potentials: Application to Lithium Diffusivity in Superionic Conducting Solid Electrolytes
• Wavelet Scattering Regression of Quantum Chemical Energies
• Representing individual electronic states for machine learning GW band structures of 2D materials
• Chemical diversity in molecular orbital energy predictions with kernel ridge regression
• Graph similarity drives zeolite diffusionless transformations and intergrowth
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Band gap prediction for large organic crystal structures with machine learning
• Navigating at Will on the Water Phase Diagram
• Hierarchical Visualization of Materials Space with Graph Convolutional Neural Networks
• Equation of state of fluid methane from first principles with machine learning potentials
• Accurate computational prediction of core-electron binding energies in carbon-based materials: A machine-learning model combining density-functional theory and $\mathbf{GW}$
• Calculation of phase diagrams in the multithermal-multibaric ensemble
• Accurate large-scale simulations of siliceous zeolites by neural network potentials
• A Bayesian approach to NMR crystal structure determination
• Phase Transitions of Zirconia: Machine-Learned Force Fields Beyond Density Functional Theory
• Machine learning force fields based on local parametrization of dispersion interactions: Application to the phase diagram of C${}_{60}$
• Examination of computed aluminum grain boundary structures and interface energies that span the 5D space of crystallographic character
• Molecular Force Fields with Gradient-Domain Machine Learning (GDML): Comparison and Synergies with Classical Force Fields
• A local order metric for condensed phase environments
• Permutationally invariant polynomial regression for energies and gradients, using reverse differentiation, achieves orders of magnitude speed-up with high precision compared to other machine learning methods
• Challenges for Machine Learning Force Fields in Reproducing Potential Energy Surfaces of Flexible Molecules
• Quantifying Chemical Structure and Atomic Energies in Amorphous Silicon Networks
• Convolutional neural networks for atomistic systems
• A Novel Approach to Describe Chemical Environments in High Dimensional Neural Network Potentials
• Prediction of Atomization Energy Using Graph Kernel and Active Learning
• $\alpha$-$\beta$ phase transition of zirconium predicted by on-the-fly machine-learned force field
• A New Kind of Atlas of Zeolite Building Blocks
• Predicting Elastic Properties of Materials from Electronic Charge Density Using 3D Deep Convolutional Neural Networks
• Generalized convex hull construction for materials discovery
• Ab initio random structure searching for battery cathode materials
• Comprehensive Analysis of the Neglect of Diatomic Differential Overlap Approximation
• Dimensionality reduction of local structure in glassy binary mixtures
• Predicting the phase diagram of titanium dioxide with random search and pattern recognition
• Unsupervised landmark analysis for jump detection in molecular dynamics simulations
• Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
• Computer Simulation of the Early Stages of Self-Assembly and Thermal Decomposition of ZIF-8
• ænet-PyTorch: a GPU-supported implementation for machine learning atomic potentials training
• Matching Crystal Structures Atom-to-Atom
• Learning with Delayed Rewards -- A case study on inverse defect design in 2D materials
• Do Machine-Learning Atomic Descriptors and Order Parameters Tell the Same Story? The Case of Liquid Water
• Recent advances in describing and driving crystal nucleation using machine learning and artificial intelligence
• Design guidelines for two-dimensional transition metal dichalcogenide alloys
• Beyond Local Structures In Critical Supercooled Water Through Unsupervised Learning
• An orbital-based representation for accurate Quantum Machine Learning
• A Physically-informed Graph-based Order Parameter for the Universal Characterization of Atomic Structures
• TimeSOAP: Tracking high-dimensional fluctuations in complex molecular systems via time-variations of SOAP spectra
• Innate Dynamics and Identity Crisis of a Metal Surface Unveiled by Machine Learning of Atomic Environments
• Structural Simplicity as a Restraint on the Structure of Amorphous Silicon
• Ephemeral ice-like local environments in classical rigid models of liquid water
• Quantifying Disorder One Atom at a Time Using an Interpretable Graph Neural Network Paradigm
• Kernel based quantum machine learning at record rate : Many-body distribution functionals as compact representations
• Visualising energy landscapes through manifold learning
• A mapping space Odyssey: characterising the statistical and metric properties of reduced representations of macromolecules
• Machine-learning accelerated identification of exfoliable two-dimensional materials
• Creation of crystal structure reproducing X-ray diffraction pattern without using database
• ZundEig: The Structure of the Proton in Liquid Water From Unsupervised Learning
• Quantum chemical roots of machine-learning molecular similarity descriptors
• Chemical Bonding in Metallic Glasses from Machine Learning and Crystal Orbital Hamilton Population
• The fundamentals of quantum machine learning
• Optimized Multifidelity Machine Learning for Quantum Chemistry
• Similarity of materials and data-quality assessment by fingerprinting
• Wasserstein metric for improved QML with adjacency matrix representations
• Automatic feature selection and weighting in molecular systems using Differentiable Information Imbalance
• Wavelet Scattering Networks for Atomistic Systems with Extrapolation of Material Properties
• Learning the energy curvature versus particle number in approximate density functionals
• Machine-learning of atomic-scale properties based on physical principles
• Mean-Field Density Matrix Decompositions
• Ab initio Canonical Sampling based on Variational Inference
• Integrated Reaction Path Processing from Sampled Structure Sequences
• Principal deuterium Hugoniot via Quantum Monte Carlo and $\Delta$-learning
• Predicting hot-electron free energies from ground-state data
• The Conformational Space of a Flexible Amino Acid at Metallic Surfaces
• Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights
• Classifying the age of a glass based on structural properties: A machine learning approach
• Local structural features elucidate crystallization of complex structures
• Inorganic Crystal Structure Prototype Database based on Unsupervised Learning of Local Atomic Environments
• Multitask methods for predicting molecular properties from heterogeneous data
• Building nonparametric $n$-body force fields using Gaussian process regression
• Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches
• Conformational analysis of tannic acid: environment effects in electronic and reactivity properties
• Metatensor and metatomic: foundational libraries for interoperable atomistic machine learning
• Machine learning at the atomic-scale
• Compressing physical properties of atomic species for improving predictive chemistry
• Machine Learning and Materials Informatics: Recent Applications and Prospects
• Cluster-based multidimensional scaling embedding tool for data visualization
• Accelerating the theoretical study of Li-polysulphide adsorption on single-atom catalysts via machine learning approaches
• Interpolating many-body wave functions for accelerated molecular dynamics on the near-exact electronic surface
• Machine Learning Framework for Modeling Exciton-Polaritons in Molecular Materials
• Reliable computational prediction of supramolecular ordering of complex molecules under electrochemical conditions
• Spin wave excitations of magnetic metalorganic materials
• Alloy Informatics through Ab Initio Charge Density Profiles: Case Study of Hydrogen Effects in Face-Centered Cubic Crystals
• Machine learning based modeling of disordered elemental semiconductors: understanding the atomic structure of a-Si and a-C
• Knowledge-Transfer based Cost-effective Search for Interface Structures: A Case Study on fcc-Al [110] Tilt Grain Boundary
• Aqueous Solution Chemistry In Silico and the Role of Data Driven Approaches
• Electronic Descriptors for Supervised Spectroscopic Predictions
• Exploring the energy landscape of aluminas through machine learning interatomic potential
• Molecular relaxation by reverse diffusion with time step prediction
• Superconductor discovery in the emerging paradigm of Materials Informatics
• Modeling refractory high-entropy alloys with efficient machine-learned interatomic potentials: defects and segregation
• Active-learning-based efficient prediction of ab-initio atomic energy: a case study on a Fe random grain boundary model with millions of atoms
• Lean CNNs for mapping electron charge density fields to material properties
• The impact of static distortion waves on superlubricity
• Dismai-Bench: Benchmarking and designing generative models using disordered materials and interfaces
• Steerable Wavelet Scattering for 3D Atomic Systems with Application to Li-Si Energy Prediction
• Self-interaction and transport of solvated electrons in molten salts
• Hybrid localized graph kernel for machine learning energy-related properties of molecules and solids
• A Fuzzy Classification Framework to Identify Equivalent Atoms in Complex Materials and Molecules
• Structural Characterization of Grain Boundaries and Machine Learning of Grain Boundary Energy and Mobility
• A Bond-Based Machine Learning Model for Molecular Polarizabilities and A Priori Raman Spectra
• Transition States Energies from Machine Learning: An Application to Reverse Water-Gas Shift on Single-Atom Alloys
• Investigating 3D Atomic Environments for Enhanced QSAR
• MADAS -- A Python framework for assessing similarity in materials-science data
• Surrogate-Based Black-Box Optimization Method for Costly Molecular Properties
• A general formalism for machine-learning models based on multipolar-spherical harmonics
• Structure-Driven Prediction of Magnetic Order in Uranium Compounds
• Point-Pattern Matching Technique for Local Structural Analysis in Condensed Matter
• Radiation damage and phase stability of Al${}_x$CrCuFeNi${}_y$ alloys using a machine-learned interatomic potential
• Prediction of Carbon Nanostructure Mechanical Properties and Role of Defects Using Machine Learning
• Orbital Mixer: Using Atomic Orbital Features for Basis Dependent Prediction of Molecular Wavefunctions
• Self-supervised Representations and Node Embedding Graph Neural Networks for Accurate and Multi-scale Analysis of Materials
• Structural transitions in dense disordered silicon from quantum-accurate ultra-large-scale simulations
• Gaussian Processes for Finite Size Extrapolation of Many-Body Simulations
• From Local Atomic Environments to Molecular Information Entropy
• Self-consistency error correction for accurate machine learning potentials from variational Monte Carlo
• Scalable Training of Neural Network Potentials for Complex Interfaces Through Data Augmentation
• Robust recognition and exploratory analysis of crystal structures via Bayesian deep learning
• Incorporating electronic information into Machine Learning potential energy surfaces via approaching the ground-state electronic energy as a function of atom-based electronic populations
• The octet rule in chemical space: Generating virtual molecules
• Training Data Set Refinement for the Machine Learning Potential of Li-Si Alloys via Structural Similarity Analysis
• Computational Design of Stable and Highly Ion-conductive Materials using Multi-objective Bayesian Optimization: Case Studies on Diffusion of Oxygen and Lithium
• Integer linear programming for unsupervised training set selection in molecular machine learning
• Efficient Data Selection Methods for the Development of Machine Learned Potentials
• Simulations of Crystal Nucleation from Solution at Constant Chemical Potential
• Using data-reduction techniques to analyse biomolecular trajectories
• A charge-density machine-learning workflow for computing the infrared spectrum of molecules
• Pairwise interactions for Potential energy surfaces and Atomic forces with Deep Neural network
• Compressing local atomic neighbourhood descriptors
• Improving accuracy of interatomic potentials: more physics or more data? A case study of silica