Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

How Well Does Kohn–Sham Regularizer Work for Weakly Correlated Systems?

Journal Article · · Journal of Physical Chemistry Letters
 [1];  [2];  [2];  [3];  [2]
  1. Univ. of California, Irvine, CA (United States); University of California, Irvine
  2. Univ. of California, Irvine, CA (United States)
  3. Google Research, Mountain View, CA (United States)
+ Show Author Affiliations
Kohn–Sham regularizer (KSR) is a differentiable machine learning approach to finding the exchange-correlation functional in Kohn–Sham density functional theory that works for strongly correlated systems. Here we test KSR for a weak correlation. We propose spin-adapted KSR (sKSR) with trainable local, semilocal, and nonlocal approximations found by minimizing density and total energy loss. We assess the atoms-to-molecules generalizability by training on one-dimensional (1D) H, He, Li, Be, and Be2+ and testing on 1D hydrogen chains, LiH, BeH2, and helium hydride complexes. The generalization error from our semilocal approximation is comparable to other differentiable approaches, but our nonlocal functional outperforms any existing machine learning functionals, predicting ground-state energies of test systems with a mean absolute error of 2.7 mH.
Research Organization:
Univ. of California, Irvine, CA (United States)
Sponsoring Organization:
National Science Foundation (NSF); USDOE Office of Science (SC)
Grant/Contract Number:
SC0008696
OSTI ID:
1866248
Journal Information:
Journal of Physical Chemistry Letters, Journal Name: Journal of Physical Chemistry Letters Journal Issue: 11 Vol. 13; ISSN 1948-7185
Publisher:
American Chemical SocietyCopyright Statement
Country of Publication:
United States
Language:
English

References (29)

Understanding machine-learned density functionals: Understanding Machine-Learned Density Functionals journal November 2015
Density Functional Theory book January 1990
Sigmoid-weighted linear units for neural network function approximation in reinforcement learning journal November 2018
Learning to Approximate Density Functionals journal February 2021
DeePKS: A Comprehensive Data-Driven Approach toward Chemically Accurate Density Functional Theory journal December 2020
Density Functional Analysis: The Theory of Density-Corrected DFT journal November 2019
Halogen and Chalcogen Binding Dominated by Density-Driven Errors journal December 2018
Bypassing the Kohn-Sham equations with machine learning journal October 2017
Machine learning accurate exchange and correlation functionals of the electronic density journal July 2020
Completing density functional theory by machine learning hidden messages from molecules journal May 2020
Array programming with NumPy journal September 2020
Reference electronic structure calculations in one dimension journal January 2012
Prescription for the design and selection of density functional approximations: More constraint satisfaction with fewer fits journal August 2005
A new local density functional for main-group thermochemistry, transition metal bonding, thermochemical kinetics, and noncovalent interactions journal November 2006
Gaussian‐2 theory for molecular energies of first‐ and second‐row compounds journal June 1991
Perspective on density functional theory journal April 2012
Thirty years of density functional theory in computational chemistry: an overview and extensive assessment of 200 density functionals journal April 2017
Inhomogeneous Electron Gas journal November 1964
Self-Consistent Equations Including Exchange and Correlation Effects journal November 1965
Highly accurate and constrained density functional obtained with differentiable programming journal October 2021
One-dimensional mimicking of electronic structure: The case for exponentials journal June 2015
Finding Density Functionals with Machine Learning journal June 2012
Understanding and Reducing Errors in Density Functional Calculations journal August 2013
Strongly Constrained and Appropriately Normed Semilocal Density Functional journal July 2015
Kohn-Sham Equations as Regularizer: Building Prior Knowledge into Machine-Learned Physics journal January 2021
Learning the Exchange-Correlation Functional from Nature with Fully Differentiable Density Functional Theory journal September 2021
Density matrix formulation for quantum renormalization groups journal November 1992
Generalized Gradient Approximation Made Simple journal October 1996
Pushing the frontiers of density functionals by solving the fractional electron problem journal December 2021

Similar Records

Machine-learning Kohn–Sham potential from dynamics in time-dependent Kohn–Sham systems
Journal Article · Sun Aug 20 20:00:00 EDT 2023 · Machine Learning: Science and Technology · OSTI ID:1996030

Related Subjects

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY