Big Data meets Quantum Chemistry Approximations: The $\Delta$-Machine Learning Approach

Abstract

Chemically accurate and comprehensive studies of the virtual space of all possible molecules are severely limited by the computational cost of quantum chemistry. We introduce a composite strategy that adds machine learning corrections to computationally inexpensive approximate legacy quantum methods. After training, highly accurate predictions of enthalpies, free energies, entropies, and electron correlation energies are possible, for significantly larger molecular sets than used for training. For thermochemical properties of up to 16k constitutional isomers of C${}_7$H${}_{10}$O${}_2$ we present numerical evidence that chemical accuracy can be reached. We also predict electron correlation energy in post Hartree-Fock methods, at the computational cost of Hartree-Fock, and we establish a qualitative relationship between molecular entropy and electron correlation. The transferability of our approach is demonstrated, using semi-empirical quantum chemistry and machine learning models trained on 1 and 10\% of 134k organic molecules, to reproduce enthalpies of all remaining molecules at density functional theory level of accuracy.

References in corpus (1)

• Nearsightedness of Electronic Matter

Cited by in corpus (173)

• SchNet - a deep learning architecture for molecules and materials
• Quantum Circuit Learning
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Less is more: sampling chemical space with active learning
• 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
• Machine learning a general purpose interatomic potential for silicon
• Machine Learning Energies of 2 M Elpasolite (ABC${}_2$D${}_6$) Crystals
• Machine learning for electronically excited states of molecules
• Prediction model of band-gap for AX binary compounds by combination of density functional theory calculations and machine learning techniques
• Alchemical and structural distribution based representation for improved QML
• The Open Catalyst 2022 (OC22) Dataset and Challenges for Oxide Electrocatalysts
• Resolving transition metal chemical space: feature selection for machine learning and structure-property relationships
• Unified Representation of Molecules and Crystals for Machine Learning
• Electronic Spectra from TDDFT and Machine Learning in Chemical Space
• Polymer Informatics: Current Status and Critical Next Steps
• OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Features
• Towards self-driving laboratories: The central role of density functional theory in the AI age
• Transferability in Machine Learning for Electronic Structure via the Molecular Orbital Basis
• Machine Learning for Quantum Mechanical Properties of Atoms in Molecules
• Incorporating long-range physics in atomic-scale machine learning
• Learning Properties of Ordered and Disordered Materials from Multi-fidelity Data
• 14 Examples of How LLMs Can Transform Materials Science and Chemistry: A Reflection on a Large Language Model Hackathon
• Conditional molecular design with deep generative models
• 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
• A Data Ecosystem to Support Machine Learning in Materials Science
• Modelling Chemical Reasoning to Predict Reactions
• Accelerated materials property predictions and design using motif-based fingerprints
• Structure-based Sampling and Self-correcting Machine Learning for Accurate Calculations of Potential Energy Surfaces and Vibrational Levels
• Machine learning of molecular properties: locality and active learning
• Machine Learning in QM/MM Molecular Dynamics Simulations of Condensed-Phase Systems
• Simulation and understanding of quantum crystals
• Quantum Machine Learning for Chemistry and Physics
• $\Delta$-Machine Learning for Potential Energy Surfaces: A PIP approach to bring a DFT-based PES to CCSD(T) Level of Theory
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via 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
• Transferable atomic multipole machine learning models for small organic molecules
• Uncovering structure-property relationships of materials by subgroup discovery
• A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
• Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces
• Machine Learning Configuration Interaction
• Machine learning based energy-free structure predictions of molecules (closed and open-shell), transition states, and solids
• A Universal Density Matrix Functional from Molecular Orbital-Based Machine Learning: Transferability across Organic Molecules
• OrbNet Denali: A machine learning potential for biological and organic chemistry with semi-empirical cost and DFT accuracy
• Uncertainty estimation for molecular dynamics and sampling
• Informing Geometric Deep Learning with Electronic Interactions to Accelerate Quantum Chemistry
• Bayesian machine learning for quantum molecular dynamics
• Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
• Breaking the Coupled Cluster Barrier for Machine Learned Potentials of Large Molecules: The Case of 15-atom Acetylacetone
• DeePKS: a comprehensive data-driven approach towards chemically accurate density functional theory
• Error-Controlled Exploration of Chemical Reaction Networks with Gaussian Processes
• Design and Analysis of Machine Learning Exchange-Correlation Functionals via Rotationally Invariant Convolutional Descriptors
• Towards the design of chemical reactions: Machine learning barriers of competing mechanisms in reactant space
• Accelerating amorphous polymer electrolyte screening by learning to reduce errors in molecular dynamics simulated properties
• Tutorial: How to Train a Neural Network Potential
• PubChemQC PM6: A dataset of 221 million molecules with optimized molecular geometries and electronic properties
• Accelerated Computation of Free Energy Profile at ab Initio Quantum Mechanical/Molecular Mechanics Accuracy via a Semi-Empirical Reference Potential. I. Weighted Thermodynamics Perturbation
• Chemical diversity in molecular orbital energy predictions with kernel ridge regression
• High-throughput property-driven generative design of functional organic molecules
• A complete description of thermodynamic stabilities of molecular crystals
• Semiempirical Molecular Orbital Models based on the Neglect of Diatomic Differential Overlap Approximation
• MLatom 3: Platform for machine learning-enhanced computational chemistry simulations and workflows
• Regression-clustering for Improved Accuracy and Training Cost with Molecular-Orbital-Based Machine Learning
• Model-data-driven constitutive responses: application to a multiscale computational framework
• Machine learning the band gap properties of kesterite I${}_2$-II-IV-V${}_4$ quaternary compounds for photovoltaics applications
• Phase Transitions of Zirconia: Machine-Learned Force Fields Beyond Density Functional Theory
• A machine learning-based selective sampling procedure for identifying the low energy region in a potential energy surface: a case study on proton conduction in oxides
• Gaussian Process-Based Refinement of Dispersion Corrections
• Universal QM/MM Approaches for General Nanoscale Applications
• The quest to simulate excited-state dynamics of transition metal complexes
• Machine Learning of Free Energies in Chemical Compound Space Using Ensemble Representations: Reaching Experimental Uncertainty for Solvation
• Putting Density Functional Theory to the Test in Machine-Learning-Accelerated Materials Discovery
• Machine Learning Prediction of Accurate Atomization Energies of Organic Molecules from Low-Fidelity Quantum Chemical Calculations
• Machine learning configuration interaction for ab initio potential energy curves
• Improved accuracy and transferability of molecular-orbital-based machine learning: Organics, transition-metal complexes, non-covalent interactions, and transition states
• Combining Machine Learning and Many-Body Calculations: Coverage-Dependent Adsorption of CO on Rh(111)
• BuRNN: Buffer Region Neural Network Approach for Polarizable-Embedding Neural Network / Molecular Mechanics Simulations
• Convolutional neural networks for atomistic systems
• Accurate machine learning force fields via experimental and simulation data fusion
• Deep Coarse-grained Potentials via Relative Entropy Minimization
• Calibrating DFT formation enthalpy calculations by multi-fidelity machine learning
• Transfer learning for chemically accurate interatomic neural network potentials
• A reactive neural network framework for water-loaded acidic zeolites
• Probabilistic performance estimators for computational chemistry methods: the empirical cumulative distribution function of absolute errors
• Machine learning the computational cost of quantum chemistry
• Structure motif centric learning framework for inorganic crystalline systems
• High pressure hydrogen by machine learning and quantum Monte Carlo
• Simulating solvation and acidity in complex mixtures with first-principles accuracy: the case of CH${}_3$SO${}_3$H and H${}_2$O${}_2$ in phenol
• ML Models of Vibrating H${}_2$CO: Comparing Reproducing Kernels, FCHL and PhysNet
• Automated construction of quantum-classical hybrid models
• Self-Parametrizing System-Focused Atomistic Models
• Steering Orbital Optimization out of Local Minima and Saddle Points Toward Lower Energy
• Molecule-specific Uncertainty Quantification in Quantum Chemical Studies
• The long road to calibrated prediction uncertainty in computational chemistry
• MLatom software ecosystem for surface hopping dynamics in Python with quantum mechanical and machine learning methods
• Scientific intuition inspired by machine learning generated hypotheses
• PhysNet Meets CHARMM: A Framework for Routine Machine Learning / Molecular Mechanics Simulations
• Impact of non-normal error distributions on the benchmarking and ranking of Quantum Machine Learning models
• Barrier height prediction by machine learning correction of semiempirical calculations
• Towards DMC accuracy across chemical space with scalable $\Delta$-QML
• Enabling robust offline active learning for machine learning potentials using simple physics-based priors
• Towards fully ab initio simulation of atmospheric aerosol nucleation
• Multi-Fidelity Machine Learning for Excited State Energies of Molecules
• An orbital-based representation for accurate Quantum Machine Learning
• Machine-learning-corrected quantum dynamics calculations
• A community-powered search of machine learning strategy space to find NMR property prediction models
• $\Delta$-Machine Learning to Elevate DFT-based Potentials and a Force Field to the CCSD(T) Level Illustrated for Ethanol
• NepoIP/MM: Towards Accurate Biomolecular Simulation with a Machine Learning/Molecular Mechanics Model Incorporating Polarization Effects
• Transfer-Learned Potential Energy Surfaces: Towards Microsecond-Scale Molecular Dynamics Simulations in the Gas Phase at CCSD(T) Quality
• A $\Delta$-Machine Learning Approach for Force Fields, Illustrated by a CCSD(T) 4-body Correction to the MB-pol Water Potential
• Probabilistic performance estimators for computational chemistry methods: Systematic Improvement Probability and Ranking Probability Matrix. I. Theory
• Resolution limit of data-driven coarse-grained models spanning chemical space
• Neural Network Potential with Multi-Resolution Approach Enables Accurate Prediction of Reaction Free Energies in Solution
• SPA${}^{\mathrm{H}}$M: the Spectrum of Approximated Hamiltonian Matrices representations
• chemtrain: Learning Deep Potential Models via Automatic Differentiation and Statistical Physics
• Critical Benchmarking of the G4(MP2) Model, the Correlation Consistent Composite Approach and Popular Density Functional Approximations on a Probabilistically Pruned Benchmark Dataset of Formation Enthalpies
• Ultra-Fast Semi-Empirical Quantum Chemistry for High-Throughput Computational Campaigns with Sparrow
• Direct Mapping Hidden Excited State Interaction Patterns from ab initio Dynamics and Its Implications on Force Field Development
• Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)
• The fundamentals of quantum machine learning
• Ab initio machine learning of phase space averages
• Optimized Multifidelity Machine Learning for Quantum Chemistry
• Quantum calculations on a new CCSD(T) machine-learned PES reveal the leaky nature of gas-phase $trans$ and $gauche$ ethanol conformers
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• High-pressure phase behaviors of titanium dioxide revealed by a $\Delta$-learning potential
• Thermal conductivity of Li${}_3$PS${}_4$ solid electrolytes with ab initio accuracy
• Assessing the persistence of chalcogen bonds in solution with neural network potentials
• Augmentation of Universal Potentials for Broad Applications
• Implicit Solvent Approach Based on Generalised Born and Transferable Graph Neural Networks for Molecular Dynamics Simulations
• Learning on-top: regressing the on-top pair density for real-space visualization of electron correlation
• Data-driven Refinement of Electronic Energies from Two-Electron Reduced-Density-Matrix Theory
• Multitask methods for predicting molecular properties from heterogeneous data
• Extending the reach of quantum computing for materials science with machine learning potentials
• Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches
• Shadow molecular dynamics and atomic cluster expansions for flexible charge models
• Training-free hyperparameter optimization of neural networks for electronic structures in matter
• Machine Learning Enhanced Calculation of Quantum-Classical Binding Free Energies
• Machine learning for accuracy in density functional approximations
• Machine learning at the atomic-scale
• Gaussian Process Regression Adaptive Density-Guided Approach: Towards Calculations of Potential Energy Surfaces for Larger Molecules
• Graph-based Quantum Response Theory and Shadow Born-Oppenheimer Molecular Dynamics
• Fortnet, a software package for training Behler-Parrinello neural networks
• Molecular Dipole Moment Learning via Rotationally Equivariant Gaussian Process Regression with Derivatives in Molecular-orbital-based Machine Learning
• High Througput pKa Prediction Using Semi Empirical Methods
• Hierarchical quantum embedding by machine learning for large molecular assemblies
• Thawed Gaussian wavepacket dynamics with $\Delta$-machine learned potentials
• Embedding quantum statistical excitations in a classical force field
• Hybrid localized graph kernel for machine learning energy-related properties of molecules and solids
• Investigating Data Hierarchies in Multifidelity Machine Learning for Excitation Energies
• All-in-one foundational models learning across quantum chemical levels
• Ab initio machine learning in chemical compound space
• Machine-learning interatomic potentials achieving CCSD(T) accuracy for systems with extended covalent networks and van der Waals interactions
• Combining Hammett $\sigma$ constants for $\Delta$-machine learning and catalyst discovery
• Band gaps of long-period polytypes of IV, IV-IV, and III-V semiconductors estimated with an Ising-type additivity model
• Assessing Non-Nested Configurations of Multifidelity Machine Learning for Quantum-Chemical Properties
• Transfer Learned Potential Energy Surfaces: Accurate Anharmonic Vibrational Dynamics and Dissociation Energies for the Formic Acid Monomer and Dimer
• Machine-Learned Potentials for Solvation Modeling
• Basis set convergence and extrapolation of connected triple excitation contributions (T) in computational thermochemistry: the W4-17 benchmark with up to k functions
• Inverse Design of Potential Singlet Fission Molecules using a Transfer Learning Based Approach
• Active Δ-learning with universal potentials for global structure optimization
• Self-consistency error correction for accurate machine learning potentials from variational Monte Carlo
• Data-Driven Modeling of S0 -> S1 Excitation Energy in the BODIPY Chemical Space: High-Throughput Computation, Quantum Machine Learning, and Inverse Design
• Gemini: Dynamic Bias Correction for Autonomous Experimentation and Molecular Simulation
• Stereo-Electronic Factors Influencing the Stability of Hydroperoxyalkyl Radicals: Transferability of Chemical Trends across Hydrocarbons and ab initio Methods
• Similarity-Based Equational Inference in Physics
• Connection between water's dynamical and structural properties: insights from ab initio simulations
• XMCQDPT2-Fidelity Transfer-Learning Potentials and a Wavepacket Oscillation Model with Power-Law Decay for Ultrafast Photodynamics
• Graph Convolutional Neural Networks for (QM)ML/MM Molecular Dynamics Simulations