December 1 - 6, 2024
Boston, Massachusetts
Symposium Supporters
2024 MRS Fall Meeting & Exhibit
MT04.12.02

Equivariant Machine Learning for Electron Density Predictions

When and Where

Dec 6, 2024
8:15am - 8:30am
Hynes, Level 2, Room 210

Presenter(s)

Co-Author(s)

Eric Taw1,2,Thomas Koker1,Keegan Quigley1,Kevin Tibbetts1,Lin Li1

Massachusetts Institute of Technology1,Lawrence Berkeley National Laboratory2

Abstract

Eric Taw1,2,Thomas Koker1,Keegan Quigley1,Kevin Tibbetts1,Lin Li1

Massachusetts Institute of Technology1,Lawrence Berkeley National Laboratory2
At the root of density functional theory, the first Hohenberg-Kohn theorem states that the ground-state properties of a material can be predicted uniquely by the electron density. We predict this fundamental quantity via an equivariant neural network called ChargE3Net and show that high rotation-order embeddings (up to l=4 spherical harmonics) can be used to accurately predict the charge density at any point in the unit cell. When trained on approximately 120,000 charge density grids from the Materials Project, we show that ChargE3Net generalizes well for bulk inorganic materials across the periodic table and accelerates self-consistent DFT calculations by 26% for non-magnetic materials when used as an initial charge density guess. Using the predicted charge densities in non-self-consistent calculations, we observe that about 40% of materials in a test set achieve <1meV/atom error for calculated potential energies. We show that ChargE3Net still accelerates DFT calculations on a subset of materials obtained from GNoME, further exemplifying its ability to generalize outside of its training set. We explore additional engineering improvements to the model and extensions to magnetic systems.

Symposium Organizers

Kjell Jorner, ETH Zurich
Jian Lin, University of Missouri-Columbia
Daniel Tabor, Texas A&M University
Dmitry Zubarev, IBM

Session Chairs

Jian Lin
Dmitry Zubarev

In this Session