Deep Learning and Density Functional Theory

Abstract

We show that deep neural networks can be integrated into, or fully replace, the Kohn-Sham density functional theory scheme for multi-electron systems in simple harmonic oscillator and random external potentials with no feature engineering. We first show that self-consistent charge densities calculated with different exchange-correlation functionals can be used as input to an extensive deep neural network to make predictions for correlation, exchange, external, kinetic and total energies simultaneously. Additionally, we show that one can also make all of the same predictions with the external potential rather than the self-consistent charge density, which allows one to circumvent the Kohn-Sham scheme altogether. We then show that a self-consistent charge density found from a non-local exchange-correlation functional can be used to make energy predictions for a semi-local exchange-correlation functional. Lastly, we use a deep convolutional inverse graphics network to predict the charge density given an external potential for different exchange-correlation functionals and assess the viability of the predicted charge densities. This work shows that extensive deep neural networks are generalizable and transferable given the variability of the potentials (maximum total energy range $\approx 100$ Ha), because they require no feature engineering, and because they can scale to an arbitrary system size with an $\mathcal{O}(N)$ computational cost.

References in corpus (15)

• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Deep Convolutional Inverse Graphics Network
• By-passing the Kohn-Sham equations with machine learning
• Finding Density Functionals with Machine Learning
• Libxc: a library of exchange and correlation functionals for density functional theory
• Real-space grids and the Octopus code as tools for the development of new simulation approaches for electronic systems
• Correlation energy and spin polarization in the 2D electron gas
• Deep learning and the Schrödinger equation
• Orbital-free Bond Breaking via Machine Learning
• Pure density functional for strong correlations and the thermodynamic limit from machine learning
• Real-space density functional theory on graphical processing units: computational approach and comparison to Gaussian basis set methods
• Extensive deep neural networks for transferring small scale learning to large scale systems
• TF.Learn: TensorFlow's High-level Module for Distributed Machine Learning
• Convolutional neural networks for atomistic systems
• Exchange-energy functionals for finite two-dimensional systems

Cited by in corpus (43)

• Opportunities and Challenges for Machine Learning in Materials Science
• Machine learning for electronically excited states of molecules
• Machine Learning in Nuclear Physics
• Quantum Machine Learning for Chemistry and Physics
• Machine Learning for Quantum Matter
• Review: Deep Learning in Electron Microscopy
• Quantum deep field: data-driven wave function, electron density generation, and atomization energy prediction and extrapolation with machine learning
• Deep learning the Hohenberg-Kohn maps of Density Functional Theory
• Toward Orbital-Free Density Functional Theory with Small Data Sets and Deep Learning
• Evolving symbolic density functionals
• Machine learning the derivative discontinuity of density-functional theory
• Unitary long-time evolution with quantum renormalization groups and artificial neural networks
• Scalable neural networks for the efficient learning of disordered quantum systems
• Twin Neural Network Regression
• Grad DFT: a software library for machine learning enhanced density functional theory
• Nuclear mass predictions based on convolutional neural network
• Unifying machine learning and quantum chemistry -- a deep neural network for molecular wavefunctions
• Machine Learning Diffusion Monte Carlo Energies
• Efficient Learning of a One-dimensional Density Functional Theory
• Calibration of nuclear charge density distribution by back-propagation neural networks
• Machine learning band gaps from the electron density
• Fast and stable deep-learning predictions of material properties for solid solution alloys
• Bridging the reality gap in quantum devices with physics-aware machine learning
• Mean-Field Density Matrix Decompositions
• A Kohn-Sham Scheme Based Neural Network for Nuclear Systems
• Transferable empirical pseudopotenials from machine learning
• Supervised learning of few dirty bosons with variable particle number
• Deep Density: circumventing the Kohn-Sham equations via symmetry preserving neural networks
• Neural networks in quantum many-body physics: a hands-on tutorial
• Deep learning density functionals for gradient descent optimization
• Insights into one-body density matrices using deep learning
• Lean CNNs for mapping electron charge density fields to material properties
• Deep learning of topological phase transitions from entanglement aspects for two-dimensional chiral p-wave superconductors
• Predicting electronic screening for fast Koopmans spectral functional calculations
• Physical machine learning outperforms "human learning" in Quantum Chemistry
• Orbital Mixer: Using Atomic Orbital Features for Basis Dependent Prediction of Molecular Wavefunctions
• Learning Density Functionals from Noisy Quantum Data
• Many-body mobility edges in 1D and 2D revealed by convolutional neural networks
• Predicting Quantum Potentials by Deep Neural Network and Metropolis Sampling
• Solving deep-learning density functional theory via variational autoencoders
• Machine-learning semi-local density functional theory for many-body lattice models at zero and finite temperature
• Electronic structures of crystalline and amorphous GeSe and GeSbTe compounds using machine learning empirical pseudopotentials
• Neural Polarization: Toward Electron Density for Molecules by Extending Equivariant Networks