Predicting the Electronic Density Response of Condensed-Phase Systems to Electric Field Perturbations

Abstract

We present a local and transferable machine learning approach capable of predicting the real-space density response of both molecules and periodic systems to external homogeneous electric fields. The new method, SALTER, builds on the Symmetry-Adapted Gaussian Process Regression SALTED framework. SALTER requires only a small, but necessary, modification to the descriptors used to represent the atomic environments. We present the performance of the method on isolated water molecules, bulk water and a naphthalene crystal. Root mean square errors of the predicted density response lie at or below 10% with barely more than 100 training structures. Derived quantities, such as polarizability tensors and even Raman spectra further derived from these tensors show a good agreement with those calculated directly from quantum mechanical methods. Therefore, SALTER shows excellent performance when predicting derived quantities, while retaining all of the information contained in the full electronic response. This method is thus capable of learning vector fields in a chemical context and serves as a landmark for further developments.

arXiv commentMain text: 7 pages, 5 figures. SI: 5 pages, 4 figures

References in corpus (23)

• Phonons and related properties of extended systems from density-functional perturbation theory
• Machine learning and the physical sciences
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges
• Machine learning for molecular simulation
• Resolution-of-identity approach to Hartree-Fock, hybrid density functionals, RPA, MP2, and \textit{GW} with numeric atom-centered orbital basis functions
• Dielectric screening in two-dimensional insulators: Implications for excitonic and impurity states in graphane
• Long-range correlation energy calculated from coupled atomic response functions
• Self-consistent hybrid functional for condensed systems
• i-PI 2.0: A Universal Force Engine for Advanced Molecular Simulations
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• A Transferable Machine-Learning Model of the Electron Density
• Accurate molecular polarizabilities with coupled-cluster theory and machine learning
• Unsupervised machine learning in atomistic simulations, between predictions and understanding
• A time-dependent density functional theory scheme for efficient calculations of dynamic (hyper)polarizabilities
• Predicting molecular dipole moments by combining atomic partial charges and atomic dipoles
• Using Gaussian Process Regression to Simulate the Vibrational Raman Spectra of Molecular Crystals
• All-Electron, Real-Space Perturbation Theory for Homogeneous Electric Fields: Theory, Implementation, and Application within DFT
• Anharmonic Effects in the Low-Frequency Vibrational Modes of Aspirin and Paracetamol Crystals
• Predicting the Charge Density Response in Metal Electrodes
• Density-functional calculation of static screening in 2D materials: the long-wavelength dielectric function of graphene
• Modeling electronic response properties with an explicit-electron machine learning potential
• Molecular Dipole Moment Learning via Rotationally Equivariant Gaussian Process Regression with Derivatives in Molecular-orbital-based Machine Learning

Cited by in corpus (4)

• Machine Learning Potential for Electrochemical Interfaces with Hybrid Representation of Dielectric Response
• Predicting the Charge Density Response in Metal Electrodes
• Learning the Electrostatic Response of the Electron Density through a Symmetry-Adapted Vector Field Model
• A charge-density machine-learning workflow for computing the infrared spectrum of molecules