Finding Density Functionals with Machine Learning

Abstract

Machine learning is used to approximate density functionals. For the model problem of the kinetic energy of non-interacting fermions in 1d, mean absolute errors below 1 kcal/mol on test densities similar to the training set are reached with fewer than 100 training densities. A predictor identifies if a test density is within the interpolation region. Via principal component analysis, a projected functional derivative finds highly accurate self-consistent densities. Challenges for application of our method to real electronic structure problems are discussed.

arXiv comment4 pages, 4 figures, 1 table. The Supplemental Material is included at the end of the manuscript (2 pages, 3 tables)

References in corpus (3)

• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Perspective on density functional theory
• Exact condition on the Kohn-Sham kinetic energy, and modern parametrization of the Thomas-Fermi density

Cited by in corpus (183)

• Methods for Interpreting and Understanding Deep Neural Networks
• Machine learning and the physical sciences
• Perspective on density functional theory
• Machine Learning Force Fields
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Comparing molecules and solids across structural and alchemical space
• Machine Learning Unifies the Modelling of Materials and Molecules
• By-passing the Kohn-Sham equations with machine learning
• Data-driven materials science: status, challenges and perspectives
• Active learning of linearly parametrized interatomic potentials
• Discovering Phase Transitions with Unsupervised Learning
• Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
• How to represent crystal structures for machine learning: towards fast prediction of electronic properties
• Opportunities and Challenges for Machine Learning in Materials Science
• SchNet: A continuous-filter convolutional neural network for modeling quantum interactions
• Autonomous discovery in the chemical sciences part I: Progress
• Resolving transition metal chemical space: feature selection for machine learning and structure-property relationships
• Representation of compounds for machine-learning prediction of physical properties
• Interatomic potentials for ionic systems with density functional accuracy based on charge densities obtained by a neural network
• Unified Representation of Molecules and Crystals for Machine Learning
• Electronic Spectra from TDDFT and Machine Learning in Chemical Space
• Accelerate Monte Carlo Simulations with Restricted Boltzmann Machines
• Towards self-driving laboratories: The central role of density functional theory in the AI age
• Accurate Interatomic Force Fields via Machine Learning with Covariant Kernels
• A Machine Learning Potential for Graphene
• Bayesian optimization for tuning and selecting hybrid-density functionals
• Perspective on integrating machine learning into computational chemistry and materials science
• Predicting Electronic Structure Properties of Transition Metal Complexes with Neural Networks
• Semilocal and Hybrid Meta-Generalized Gradient Approximations Based on the Understanding of the Kinetic-Energy-Density Dependence
• DFT: A Theory Full of Holes?
• Kohn-Sham equations as regularizer: building prior knowledge into machine-learned physics
• Deep learning and the Schrödinger equation
• Quantum Machine Learning for Chemistry and Physics
• Machine Learning for Quantum Matter
• Machine learning for many-body physics: The case of the Anderson impurity model
• Orbital-free Bond Breaking via Machine Learning
• The Importance of being consistent
• Deep Learning and Density Functional Theory
• Machine learning and density functional theory
• A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
• Pure density functional for strong correlations and the thermodynamic limit from machine learning
• Machine Learning the Physical Non-Local Exchange-Correlation Functional of Density-Functional Theory
• The Many-Body Expansion Combined with Neural Networks
• Advances of Machine Learning in Materials Science: Ideas and Techniques
• Learning the exchange-correlation functional from nature with fully differentiable density functional theory
• Predicting the Curie temperature of ferromagnets using machine learning
• Electronic Density Response of Warm Dense Matter
• Semilocal Pauli-Gaussian Kinetic Functionals for Orbital-Free Density Functional Theory Calculations of Solids
• Supervised Learning with Quantum-Inspired Tensor Networks
• DeePKS: a comprehensive data-driven approach towards chemically accurate density functional theory
• Machine Learning for Condensed Matter Physics
• Design and Analysis of Machine Learning Exchange-Correlation Functionals via Rotationally Invariant Convolutional Descriptors
• Machine-learning inference of fluid variables from data using reservoir computing
• Laplacian-level kinetic energy approximations based on the fourth-order gradient expansion: Global assessment and application to the subsystem formulation of density functional theory
• Metadynamics for Training Neural Network Model Chemistries: a Competitive Assessment
• PubChemQC PM6: A dataset of 221 million molecules with optimized molecular geometries and electronic properties
• Thermodynamics and its Prediction and CALPHAD Modeling: Review, State of the Art, and Perspectives
• Neural-network Kohn-Sham exchange-correlation potential and its out-of-training transferability
• Identifying polymer states by machine learning
• KITE: high-performance accurate modelling of electronic structure and response functions of large molecules, disordered crystals and heterostructures
• Modeling Electronic Quantum Transport with Machine Learning
• Pattern Learning Electronic Density of States
• Compositional optimization of hard-magnetic phases with machine-learning models
• Deep learning the Hohenberg-Kohn maps of Density Functional Theory
• Accelerating Finite-temperature Kohn-Sham Density Functional Theory with Deep Neural Networks
• Nuclear energy density functionals from machine learning
• Laplacian-dependent models of the kinetic energy density: Applications in subsystem density functional theory with meta-generalized gradient approximation functionals
• Non-local kinetic energy functional from the Jellium-with-gap model: applications to Orbital-Free Density Functional Theory
• Group-theoretical high-order rotational invariants for structural representations: Application to linearized machine learning interatomic potential
• Thermal Density Functional Theory in Context
• A Novel Approach to Describe Chemical Environments in High Dimensional Neural Network Potentials
• A classical density functional from machine learning and a convolutional neural network
• Toward Orbital-Free Density Functional Theory with Small Data Sets and Deep Learning
• Variational quantum amplitude estimation
• Data-driven kinetic energy density fitting for orbital-free DFT: linear vs Gaussian process regression
• Quantum Machine-Learning for Eigenstate Filtration in Two-Dimensional Materials
• Evolving symbolic density functionals
• Developments and Further Applications of Ephemeral Data Derived Potentials
• Machine learning \& artificial intelligence in the quantum domain
• Analytical classical density functionals from an equation learning network
• Overcoming the Barrier of Orbital-Free Density Functional Theory for Molecular Systems Using Deep Learning
• Structure motif centric learning framework for inorganic crystalline systems
• A machine-learning solver for modified diffusion equations
• Exploring a potential energy surface by machine learning for characterizing atomic transport
• CIDER: An Expressive, Nonlocal Feature Set for Machine Learning Density Functionals with Exact Constraints
• Machine learning based localization and classification with atomic magnetometers
• Learning from the Density to Correct Total Energy and Forces in First Principle Simulations
• First-Principles Prediction of Electronic Transport in Experimental Semiconductor Heterostructures via Physics-Based Machine Learning
• Machine Learning Exchange-Correlation Potential in Time Dependent Density Functional Theory
• Order-$N$ orbital-free density-functional calculations with machine learning of functional derivatives for semiconductors and metals
• Genarris: Random Generation of Molecular Crystal Structures and Fast Screening with a Harris Approximation
• Optimization of an exchange-correlation density functional for water
• Machine learning of a density functional for anisotropic patchy particles
• Deep Learning the Functional Renormalization Group
• Multi-Fidelity Machine Learning for Excited State Energies of Molecules
• Protocol Discovery for the Quantum Control of Majoranas by Differentiable Programming and Natural Evolution Strategies
• Supervised machine learning of ultracold atoms with speckle disorder
• Machine learning electron correlation in a disordered medium
• Scalable neural networks for the efficient learning of disordered quantum systems
• Linear Jacobi-Legendre expansion of the charge density for machine learning-accelerated electronic structure calculations
• Method for Estimating Spin-Spin Interactions from Magnetization Curves
• Reliable emulation of complex functionals by active learning with error control
• A Neural Networks study of the phase transitions of Potts model
• Emergent Quantum Mechanics in an Introspective Machine Learning Architecture
• Machine learning nonequilibrium electron forces for adiabatic spin dynamics
• Exact constraints and appropriate norms in machine learned exchange-correlation functionals
• Kernel based quantum machine learning at record rate : Many-body distribution functionals as compact representations
• Machine learning density functionals from the random-phase approximation
• Systematic construction of density functionals based on matrix product state computations
• Tree based machine learning framework for predicting ground state energies of molecules
• Grad DFT: a software library for machine learning enhanced density functional theory
• Machine Learning Diffusion Monte Carlo Energies
• Efficient Learning of a One-dimensional Density Functional Theory
• Machine learning effective models for quantum systems
• DiracSolver: a tool for solving the Dirac Equation
• Bridging electronic and classical density-functional theory using universal machine-learned functional approximations
• Completing density functional theory by machine-learning hidden messages from molecules
• Sampling algorithms for validation of supervised learning models for Ising-like systems
• Artificial Intelligence for Science in Quantum, Atomistic, and Continuum Systems
• A transferable artificial neural network model for atomic forces in nanoparticles
• Reducing the Quantum Many-electron Problem to Two Electrons with Machine Learning
• Variational principle to regularize machine-learned density functionals: the non-interacting kinetic-energy functional
• Mean-Field Density Matrix Decompositions
• Kinetic Energy Density Functionals by Axiomatic Approach
• Addressing the Band Gap Problem with a Machine-Learned Exchange Functional
• Stochastic Replica Voting Machine Prediction of Stable Cubic and Double Perovskite Materials and Binary Alloys
• Density functionals and Kohn-Sham potentials with minimal wavefunction preparations on a quantum computer
• Transferable empirical pseudopotenials from machine learning
• Machine learning phases and criticalities without using real data for training
• Supervised learning of few dirty bosons with variable particle number
• The loss of the property of locality of the kernel in high-dimensional Gaussian process regression on the example of the fitting of molecular potential energy surfaces
• Machine learning based nonlocal kinetic energy density functional for simple metals and alloys
• Machine learning forces trained by Gaussian process in liquid states: Transferability to temperature and pressure
• How Well Does Kohn-Sham Regularizer Work for Weakly Correlated Systems?
• Neural networks in quantum many-body physics: a hands-on tutorial
• (Spin-)density-functional theory for open-shell systems: exact magnetization density functional for the half-filled Hubbard trimer
• Training-free hyperparameter optimization of neural networks for electronic structures in matter
• Machine Learning and Materials Informatics: Recent Applications and Prospects
• Generation of Bose-Einstein Condensates' Ground State Through Machine Learning
• Multi-channel machine learning based nonlocal kinetic energy density functional for semiconductors
• Nonequilibrium fluctuations of a driven quantum heat engine via machine learning
• Machine learning for accuracy in density functional approximations
• Levy-Lieb principle: The bridge between the electron density of Density Functional Theory and the wavefunction of Quantum Monte Carlo
• Neural-network-supported basis optimizer for the configuration interaction problem in quantum many-body clusters: Feasibility study and numerical proof
• Variational Autoencoder Analysis of Ising Model Statistical Distributions and Phase Transitions
• Machine learning phases of an Abelian gauge theory
• Detecting Nematic Order in STM/STS Data with Artificial Intelligence
• Deep learning density functionals for gradient descent optimization
• Decoupling approximation robustly reconstructs directed dynamical networks
• Insights into one-body density matrices using deep learning
• Conservation Laws and Spin System Modeling through Principal Component Analysis
• Deep learning nonlocal and scalable energy functionals for quantum Ising models
• A comparison of three ways to measure time-dependent densities with quantum simulators
• Discrete Fourier Transform Improves the Prediction of the Electronic Properties of Molecules in Quantum Machine Learning
• Ab initio machine learning in chemical compound space
• Self-consistent assessment of Englert-Schwinger model on atomic properties
• Physical machine learning outperforms "human learning" in Quantum Chemistry
• Solution of inverse problem for Gross-Pitaevskii equation with artificial neural networks
• On the Kinetic Energy Density Functional: The Limit of the Density Derivative Order
• Efficient Mean-Field Simulation of Quantum Circuits Inspired by Density Functional Theory
• Orbital Mixer: Using Atomic Orbital Features for Basis Dependent Prediction of Molecular Wavefunctions
• Learning Density Functionals from Noisy Quantum Data
• Flexible dual-branched message passing neural network for quantum mechanical property prediction with molecular conformation
• Learning DFT
• Bridging microscopy with molecular dynamics and quantum simulations: An AtomAI based pipeline
• Algorithmic differentiation for plane-wave DFT: materials design, error control and learning model parameters
• Incorporating electronic information into Machine Learning potential energy surfaces via approaching the ground-state electronic energy as a function of atom-based electronic populations
• Machine-learning semi-local density functional theory for many-body lattice models at zero and finite temperature
• Predicting Many Crystal Properties via an Adaptive Transformer-based Framework
• Solving deep-learning density functional theory via variational autoencoders
• atoMEC: An open-source average-atom Python code
• Machine learning reveals orbital interaction in crystalline materials
• Graph Neural Network for Hamiltonian-Based Material Property Prediction
• Global sampling of Feynman's diagrams through Normalizing Flow
• Machine learning approach for vibronically renormalized electronic band structures
• Neural network distillation of orbital dependent density functional theory
• Exchange-Correlation Functionals in 2D Materials: Applications, Challenges, and Limitations
• Learning Minimal Representations of Fermionic Ground States
• Challenges and opportunities in the supervised learning of quantum circuit outputs
• Reconstruction of observed mechanical motions with Artificial Intelligence tools
• HOAX: A Hyperparameter Optimization Algorithm Explorer for Neural Networks