Understanding molecular representations in machine learning: The role of uniqueness and target similarity

Abstract

The predictive accuracy of Machine Learning (ML) models of molecular properties depends on the choice of the molecular representation. Based on the postulates of quantum mechanics, we introduce a hierarchy of representations which meet uniqueness and target similarity criteria. To systematically control target similarity, we rely on interatomic many body expansions, as implemented in universal force-fields, including Bonding, Angular, and higher order terms (BA). Addition of higher order contributions systematically increases similarity to the true potential energy and predictive accuracy of the resulting ML models. We report numerical evidence for the performance of BAML models trained on molecular properties pre-calculated at electron-correlated and density functional theory level of theory for thousands of small organic molecules. Properties studied include enthalpies and free energies of atomization, heatcapacity, zero-point vibrational energies, dipole-moment, polarizability, HOMO/LUMO energies and gap, ionization potential, electron affinity, and electronic excitations. After training, BAML predicts energies or electronic properties of out-of-sample molecules with unprecedented accuracy and speed.

References in corpus (9)

• On representing chemical environments
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Big Data of Materials Science - Critical Role of the Descriptor
• Comparing molecules and solids across structural and alchemical space
• Machine Learning of Molecular Electronic Properties in Chemical Compound Space
• How to represent crystal structures for machine learning: towards fast prediction of electronic properties
• Machine Learning Energies of 2 M Elpasolite (ABC${}_2$D${}_6$) Crystals
• Wavelet Scattering Regression of Quantum Chemical Energies
• Toward transferable interatomic van der Waals interactions without electrons: The role of multipole electrostatics and many-body dispersion

Cited by in corpus (71)

• Machine Learning Force Fields
• Molecular Contrastive Learning of Representations via Graph Neural Networks
• Machine Learning Unifies the Modelling of Materials and Molecules
• Machine learning prediction errors better than DFT accuracy
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
• Physics-inspired structural representations for molecules and materials
• Opportunities and Challenges for Machine Learning in Materials Science
• Machine learning a general purpose interatomic potential for silicon
• Machine learning for electronically excited states of molecules
• ANI-1: A data set of 20M off-equilibrium DFT calculations for organic molecules
• 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
• Automatic Selection of Atomic Fingerprints and Reference Configurations for Machine-Learning Potentials
• WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
• Unified Representation of Molecules and Crystals for Machine Learning
• Applying machine learning techniques to predict the properties of energetic materials
• Towards self-driving laboratories: The central role of density functional theory in the AI age
• On the Completeness of Atomic Structure Representations
• Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks
• Machine learning with force-field inspired descriptors for materials: fast screening and mapping energy landscape
• Unsupervised machine learning in atomistic simulations, between predictions and understanding
• Atom-Density Representations for Machine Learning
• Non-covalent interactions across organic and biological subsets of chemical space: Physics-based potentials parametrized from machine learning
• Machine learning of molecular properties: locality and active learning
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Efficient nonparametric $n$-body force fields from machine learning
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements
• Neural Network Potentials for Chemistry: Concepts, Applications and Prospects
• Operators in Machine Learning: Response Properties in Chemical Space
• Machine learning based energy-free structure predictions of molecules (closed and open-shell), transition states, and solids
• Accelerating CALYPSO Structure Prediction by Data-driven Learning of Potential Energy Surface
• Maximizing information from chemical engineering data sets: Applications to machine learning
• A review of mathematical representations of biomolecules
• Chemical diversity in molecular orbital energy predictions with kernel ridge regression
• PiNN: A Python Library for Building Atomic Neural Networks of Molecules and Materials
• Recursive evaluation and iterative contraction of $N$-body equivariant features
• Deep Learning for UV Absorption Spectra with SchNarc: First Steps Towards Transferability in Chemical Compound Space
• Machine learning and excited-state molecular dynamics
• Using Molecular Embeddings in QSAR Modeling: Does it Make a Difference?
• Machine learning enhanced global optimization by clustering local environments to enable bundled atomic energies
• Global optimization of atomic structures with gradient-enhanced Gaussian process regression
• Prediction of Atomization Energy Using Graph Kernel and Active Learning
• Machine learning the computational cost of quantum chemistry
• Atomistic structure learning
• A New Kind of Atlas of Zeolite Building Blocks
• Comprehensive Analysis of the Neglect of Diatomic Differential Overlap Approximation
• Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning
• Symmetrical Graph Neural Network for Quantum Chemistry, with Dual R/K Space
• Genarris: Random Generation of Molecular Crystal Structures and Fast Screening with a Harris Approximation
• Towards DMC accuracy across chemical space with scalable $\Delta$-QML
• An orbital-based representation for accurate Quantum Machine Learning
• Controlled exploration of chemical space by machine learning of coarse-grained representations
• Kernel based quantum machine learning at record rate : Many-body distribution functionals as compact representations
• Optimal radial basis for density-based atomic representations
• SPA${}^{\mathrm{H}}$M: the Spectrum of Approximated Hamiltonian Matrices representations
• Size and Temperature Transferability of Direct and Local Deep Neural Networks for Atomic Forces
• Quantum chemical roots of machine-learning molecular similarity descriptors
• Ab initio machine learning of phase space averages
• The fundamentals of quantum machine learning
• Wasserstein metric for improved QML with adjacency matrix representations
• Learning the energy curvature versus particle number in approximate density functionals
• Decomposing Chemical Space: Applications to the Machine Learning of Atomic Energies
• Mean-Field Density Matrix Decompositions
• Supervised learning of few dirty bosons with variable particle number
• Generating candidates in global optimization algorithms using complementary energy landscapes
• Compressing physical properties of atomic species for improving predictive chemistry
• 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"
• Electronic Descriptors for Supervised Spectroscopic Predictions
• Molecular Hessian matrices from a machine learning random forest regression algorithm