How Well Does Kohn–Sham Regularizer Work for Weakly Correlated Systems?
Abstract
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.
- Authors:
-
- Univ. of California, Irvine, CA (United States)
- Google Research, Mountain View, CA (United States)
- Publication Date:
- Research Org.:
- Univ. of California, Irvine, CA (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC); National Science Foundation (NSF)
- OSTI Identifier:
- 1866248
- Grant/Contract Number:
- SC0008696; CHE-1856165; DGE-1633631
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Journal of Physical Chemistry Letters
- Additional Journal Information:
- Journal Volume: 13; Journal Issue: 11; Journal ID: ISSN 1948-7185
- Publisher:
- American Chemical Society
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY
Citation Formats
Kalita, Bhupalee, Pederson, Ryan, Chen, Jielun, Li, Li, and Burke, Kieron. How Well Does Kohn–Sham Regularizer Work for Weakly Correlated Systems?. United States: N. p., 2022.
Web. doi:10.1021/acs.jpclett.2c00371.
Kalita, Bhupalee, Pederson, Ryan, Chen, Jielun, Li, Li, & Burke, Kieron. How Well Does Kohn–Sham Regularizer Work for Weakly Correlated Systems?. United States. https://doi.org/10.1021/acs.jpclett.2c00371
Kalita, Bhupalee, Pederson, Ryan, Chen, Jielun, Li, Li, and Burke, Kieron. Mon .
"How Well Does Kohn–Sham Regularizer Work for Weakly Correlated Systems?". United States. https://doi.org/10.1021/acs.jpclett.2c00371. https://www.osti.gov/servlets/purl/1866248.
@article{osti_1866248,
title = {How Well Does Kohn–Sham Regularizer Work for Weakly Correlated Systems?},
author = {Kalita, Bhupalee and Pederson, Ryan and Chen, Jielun and Li, Li and Burke, Kieron},
abstractNote = {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.},
doi = {10.1021/acs.jpclett.2c00371},
journal = {Journal of Physical Chemistry Letters},
number = 11,
volume = 13,
place = {United States},
year = {Mon Mar 14 00:00:00 EDT 2022},
month = {Mon Mar 14 00:00:00 EDT 2022}
}
Works referenced in this record:
Inhomogeneous Electron Gas
journal, November 1964
- Hohenberg, P.; Kohn, W.
- Physical Review, Vol. 136, Issue 3B, p. B864-B871
Self-Consistent Equations Including Exchange and Correlation Effects
journal, November 1965
- Kohn, W.; Sham, L. J.
- Physical Review, Vol. 140, Issue 4A, p. A1133-A1138
Thirty years of density functional theory in computational chemistry: an overview and extensive assessment of 200 density functionals
journal, April 2017
- Mardirossian, Narbe; Head-Gordon, Martin
- Molecular Physics, Vol. 115, Issue 19
Generalized Gradient Approximation Made Simple
journal, October 1996
- Perdew, John P.; Burke, Kieron; Ernzerhof, Matthias
- Physical Review Letters, Vol. 77, Issue 18, p. 3865-3868
A new local density functional for main-group thermochemistry, transition metal bonding, thermochemical kinetics, and noncovalent interactions
journal, November 2006
- Zhao, Yan; Truhlar, Donald G.
- The Journal of Chemical Physics, Vol. 125, Issue 19, Article No. 194101
Perspective on density functional theory
journal, April 2012
- Burke, Kieron
- The Journal of Chemical Physics, Vol. 136, Issue 15
Understanding machine-learned density functionals: Understanding Machine-Learned Density Functionals
journal, November 2015
- Li, Li; Snyder, John C.; Pelaschier, Isabelle M.
- International Journal of Quantum Chemistry, Vol. 116, Issue 11
Bypassing the Kohn-Sham equations with machine learning
journal, October 2017
- Brockherde, Felix; Vogt, Leslie; Li, Li
- Nature Communications, Vol. 8, Issue 1
Learning to Approximate Density Functionals
journal, February 2021
- Kalita, Bhupalee; Li, Li; McCarty, Ryan J.
- Accounts of Chemical Research, Vol. 54, Issue 4
Completing density functional theory by machine learning hidden messages from molecules
journal, May 2020
- Nagai, Ryo; Akashi, Ryosuke; Sugino, Osamu
- npj Computational Materials, Vol. 6, Issue 1
Gaussian‐2 theory for molecular energies of first‐ and second‐row compounds
journal, June 1991
- Curtiss, Larry A.; Raghavachari, Krishnan; Trucks, Gary W.
- The Journal of Chemical Physics, Vol. 94, Issue 11
Understanding and Reducing Errors in Density Functional Calculations
journal, August 2013
- Kim, Min-Cheol; Sim, Eunji; Burke, Kieron
- Physical Review Letters, Vol. 111, Issue 7
Kohn-Sham Equations as Regularizer: Building Prior Knowledge into Machine-Learned Physics
journal, January 2021
- Li, Li; Hoyer, Stephan; Pederson, Ryan
- Physical Review Letters, Vol. 126, Issue 3
Array programming with NumPy
journal, September 2020
- Harris, Charles R.; Millman, K. Jarrod; van der Walt, Stéfan J.
- Nature, Vol. 585, Issue 7825
Learning the Exchange-Correlation Functional from Nature with Fully Differentiable Density Functional Theory
journal, September 2021
- Kasim, M. F.; Vinko, S. M.
- Physical Review Letters, Vol. 127, Issue 12
Strongly Constrained and Appropriately Normed Semilocal Density Functional
journal, July 2015
- Sun, Jianwei; Ruzsinszky, Adrienn; Perdew, John P.
- Physical Review Letters, Vol. 115, Issue 3
Machine learning accurate exchange and correlation functionals of the electronic density
journal, July 2020
- Dick, Sebastian; Fernandez-Serra, Marivi
- Nature Communications, Vol. 11, Issue 1
DeePKS: A Comprehensive Data-Driven Approach toward Chemically Accurate Density Functional Theory
journal, December 2020
- Chen, Yixiao; Zhang, Linfeng; Wang, Han
- Journal of Chemical Theory and Computation, Vol. 17, Issue 1
Pushing the frontiers of density functionals by solving the fractional electron problem
journal, December 2021
- Kirkpatrick, James; McMorrow, Brendan; Turban, David H. P.
- Science, Vol. 374, Issue 6573
Prescription for the design and selection of density functional approximations: More constraint satisfaction with fewer fits
journal, August 2005
- Perdew, John P.; Ruzsinszky, Adrienn; Tao, Jianmin
- The Journal of Chemical Physics, Vol. 123, Issue 6
Density matrix formulation for quantum renormalization groups
journal, November 1992
- White, Steven R.
- Physical Review Letters, Vol. 69, Issue 19
Sigmoid-weighted linear units for neural network function approximation in reinforcement learning
journal, November 2018
- Elfwing, Stefan; Uchibe, Eiji; Doya, Kenji
- Neural Networks, Vol. 107
Density Functional Analysis: The Theory of Density-Corrected DFT
journal, November 2019
- Vuckovic, Stefan; Song, Suhwan; Kozlowski, John
- Journal of Chemical Theory and Computation, Vol. 15, Issue 12
Halogen and Chalcogen Binding Dominated by Density-Driven Errors
journal, December 2018
- Kim, Yeil; Song, Suhwan; Sim, Eunji
- The Journal of Physical Chemistry Letters, Vol. 10, Issue 2
Reference electronic structure calculations in one dimension
journal, January 2012
- Wagner, Lucas O.; Stoudenmire, E. M.; Burke, Kieron
- Physical Chemistry Chemical Physics, Vol. 14, Issue 24