Training models using forces computed by stochastic electronic structure methods

Ceperley, David M.; Jensen, Scott; Yang, Yubo; Niu, Hongwei; Pierleoni, Carlo; Holzmann, Markus

doi:10.1088/2516-1075/ad2eb0

Title: Training models using forces computed by stochastic electronic structure methods

Journal Article · Fri Mar 15 00:00:00 EDT 2024 · Electronic Structure

DOI:https://doi.org/10.1088/2516-1075/ad2eb0· OSTI ID:2323949

;

; Niu, Hongwei;

; Holzmann, Markus

Abstract Quantum Monte Carlo (QMC) can play a very important role in generating accurate data needed for constructing potential energy surfaces. We argue that QMC has advantages in terms of a smaller systematic bias and an ability to cover phase space more completely. The stochastic noise can ease the training of the machine learning model. We discuss how stochastic errors affect the generation of effective models by analyzing the errors within a linear least squares procedure, finding that there is an advantage to having many relatively imprecise data points for constructing models. We then analyze the effect of noise on a model of many-body silicon finding that noise in some situations improves the resulting model. We then study the effect of QMC noise on two machine learning models of dense hydrogen used in a recent study of its phase diagram. The noise enables us to estimate the errors in the model. We conclude with a discussion of future research problems.

View Journal Article

Cite

Export

Save

Sponsoring Organization:: USDOE Office of Science (SC), Basic Energy Sciences (BES)

Grant/Contract Number:: DOE DE-SC0020177

OSTI ID:: 2323949

Alternate ID(s):: OSTI ID: 2316069

Journal Information:: Electronic Structure, Journal Name: Electronic Structure Vol. 6 Journal Issue: 1; ISSN 2516-1075

Publisher:: IOP PublishingCopyright Statement

Country of Publication:: United Kingdom

Language:: English

References (40)

Generalized Gradient Approximation Made Simple Perdew, John P.; Burke, Kieron; Ernzerhof, Matthias Physical Review Letters, Vol. 77, Issue 18, p. 3865-3868 https://doi.org/10.1103/PhysRevLett.77.3865	journal	October 1996
New stochastic method for systems with broken time-reversal symmetry: 2D fermions in a magnetic field Ortiz, G.; Ceperley, D. M.; Martin, R. M. Physical Review Letters, Vol. 71, Issue 17 https://doi.org/10.1103/PhysRevLett.71.2777	journal	October 1993
Gaussian Process Regression for Materials and Molecules Deringer, Volker L.; Bartók, Albert P.; Bernstein, Noam Chemical Reviews, Vol. 121, Issue 16 https://doi.org/10.1021/acs.chemrev.1c00022	journal	August 2021
Benchmarking density functionals for hydrogen-helium mixtures with quantum Monte Carlo: Energetics, pressures, and forces Clay, Raymond C.; Holzmann, Markus; Ceperley, David M. Physical Review B, Vol. 93, Issue 3 https://doi.org/10.1103/PhysRevB.93.035121	journal	January 2016
Space-warp coordinate transformation for efficient ionic force calculations in quantum Monte Carlo Nakano, Kousuke; Raghav, Abhishek; Sorella, Sandro The Journal of Chemical Physics, Vol. 156, Issue 3 https://doi.org/10.1063/5.0076302	journal	January 2022
High-pressure hydrogen by machine learning and quantum Monte Carlo Tirelli, Andrea; Tenti, Giacomo; Nakano, Kousuke Physical Review B, Vol. 106, Issue 4 https://doi.org/10.1103/PhysRevB.106.L041105	journal	July 2022
First Principles Methods: A Perspective from Quantum Monte Carlo Morales, Miguel; Clay, Raymond; Pierleoni, Carlo Entropy, Vol. 16, Issue 1 https://doi.org/10.3390/e16010287	journal	December 2013
Forces in Molecules Feynman, R. P. Physical Review, Vol. 56, Issue 4 https://doi.org/10.1103/PhysRev.56.340	journal	August 1939
Neural network ansatz for periodic wave functions and the homogeneous electron gas Wilson, Max; Moroni, Saverio; Holzmann, Markus Physical Review B, Vol. 107, Issue 23 https://doi.org/10.1103/PhysRevB.107.235139	journal	June 2023
Neural-network quantum states for periodic systems in continuous space Pescia, Gabriel; Han, Jiequn; Lovato, Alessandro Physical Review Research, Vol. 4, Issue 2 https://doi.org/10.1103/PhysRevResearch.4.023138	journal	May 2022
Variational and diffusion quantum Monte Carlo calculations with the CASINO code Needs, R. J.; Towler, M. D.; Drummond, N. D. The Journal of Chemical Physics, Vol. 152, Issue 15 https://doi.org/10.1063/1.5144288	journal	April 2020
Machine Learning Diffusion Monte Carlo Forces Huang, Cancan; Rubenstein, Brenda M. The Journal of Physical Chemistry A, Vol. 127, Issue 1 https://doi.org/10.1021/acs.jpca.2c05904	journal	December 2022
Nonlinear Network Description for Many-Body Quantum Systems in Continuous Space Ruggeri, Michele; Moroni, Saverio; Holzmann, Markus Physical Review Letters, Vol. 120, Issue 20 https://doi.org/10.1103/PhysRevLett.120.205302	journal	May 2018
Towards the Solution of the Many-Electron Problem in Real Materials: Equation of State of the Hydrogen Chain with State-of-the-Art Many-Body Methods Motta, Mario; Ceperley, David M.; Chan, Garnet Kin-Lic Physical Review X, Vol. 7, Issue 3 https://doi.org/10.1103/PhysRevX.7.031059	journal	September 2017
Evidence for supercritical behaviour of high-pressure liquid hydrogen Cheng, Bingqing; Mazzola, Guglielmo; Pickard, Chris J. Nature, Vol. 585, Issue 7824 https://doi.org/10.1038/s41586-020-2677-y	journal	September 2020
Accurate Computation of Vibronic Energies and of Some Expectation Values for H ₂ , D ₂ , and T ₂ Kol/os, W.; Wolniewicz, L. The Journal of Chemical Physics, Vol. 41, Issue 12 https://doi.org/10.1063/1.1725797	journal	December 1964
Communication: Energy benchmarking with quantum Monte Carlo for water nano-droplets and bulk liquid water Alfè, D.; Bartók, A. P.; Csányi, G. The Journal of Chemical Physics, Vol. 138, Issue 22 https://doi.org/10.1063/1.4810882	journal	June 2013
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces Behler, Jörg; Parrinello, Michele Physical Review Letters, Vol. 98, Issue 14 https://doi.org/10.1103/PhysRevLett.98.146401	journal	April 2007
Deep Variational Free Energy Approach to Dense Hydrogen Xie, Hao; Li, Zi-Hang; Wang, Han Physical Review Letters, Vol. 131, Issue 12 https://doi.org/10.1103/PhysRevLett.131.126501	journal	September 2023
Trial wave functions for high-pressure metallic hydrogen Pierleoni, Carlo; Delaney, Kris T.; Morales, Miguel A. Computer Physics Communications, Vol. 179, Issue 1-3 https://doi.org/10.1016/j.cpc.2008.01.041	journal	July 2008
QMCPACK: Advances in the development, efficiency, and application of auxiliary field and real-space variational and diffusion quantum Monte Carlo Kent, P. R. C.; Annaberdiyev, Abdulgani; Benali, Anouar The Journal of Chemical Physics, Vol. 152, Issue 17 https://doi.org/10.1063/5.0004860	journal	May 2020
TurboRVB : A many-body toolkit for ab initio electronic simulations by quantum Monte Carlo Nakano, Kousuke; Attaccalite, Claudio; Barborini, Matteo The Journal of Chemical Physics, Vol. 152, Issue 20 https://doi.org/10.1063/5.0005037	journal	May 2020
The properties of hydrogen and helium under extreme conditions McMahon, Jeffrey M.; Morales, Miguel A.; Pierleoni, Carlo Reviews of Modern Physics, Vol. 84, Issue 4 https://doi.org/10.1103/RevModPhys.84.1607	journal	November 2012
Ground State of the Electron Gas by a Stochastic Method Ceperley, D. M.; Alder, B. J. Physical Review Letters, Vol. 45, Issue 7, p. 566-569 https://doi.org/10.1103/PhysRevLett.45.566	journal	August 1980
Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons Bartók, Albert P.; Payne, Mike C.; Kondor, Risi Physical Review Letters, Vol. 104, Issue 13 https://doi.org/10.1103/PhysRevLett.104.136403	journal	April 2010
Ab-initio Study of Interacting Fermions at Finite Temperature with Neural Canonical Transformation Xie, Hao; null, Linfeng Zhang; Wang, Lei Journal of Machine Learning, Vol. 1, Issue 1 https://doi.org/10.4208/jml.220113	journal	June 2022
LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales Thompson, Aidan P.; Aktulga, H. Metin; Berger, Richard Computer Physics Communications, Vol. 271 https://doi.org/10.1016/j.cpc.2021.108171	journal	February 2022
Twist-averaged boundary conditions in continuum quantum Monte Carlo algorithms Lin, C.; Zong, F. H.; Ceperley, D. M. Physical Review E, Vol. 64, Issue 1 https://doi.org/10.1103/PhysRevE.64.016702	journal	June 2001
Closed-loop direct control of seizure focus in a rodent model of temporal lobe epilepsy via localized electric fields applied sequentially Kang, Wonok; Ju, Chanyang; Joo, Jaesoon Nature Communications, Vol. 13, Issue 1 https://doi.org/10.1038/s41467-022-35540-7	journal	December 2022
Computational Methods in Coupled Electron-Ion Monte Carlo Simulations Pierleoni, Carlo; Ceperley, David M. ChemPhysChem, Vol. 6, Issue 9 https://doi.org/10.1002/cphc.200400587	journal	September 2005
`QMCPACK` : an open source ab initio quantum Monte Carlo package for the electronic structure of atoms, molecules and solids Kim, Jeongnim; Baczewski, Andrew D.; Beaudet, Todd D. Journal of Physics: Condensed Matter, Vol. 30, Issue 19 https://doi.org/10.1088/1361-648X/aab9c3	journal	April 2018
Unified Approach for Molecular Dynamics and Density-Functional Theory Car, R.; Parrinello, M. Physical Review Letters, Vol. 55, Issue 22 https://doi.org/10.1103/PhysRevLett.55.2471	journal	November 1985
Quantum Monte Carlo for molecules: Green’s function and nodal release Ceperley, D. M.; Alder, B. J. The Journal of Chemical Physics, Vol. 81, Issue 12 https://doi.org/10.1063/1.447637	journal	December 1984
New empirical approach for the structure and energy of covalent systems Tersoff, J. Physical Review B, Vol. 37, Issue 12 https://doi.org/10.1103/PhysRevB.37.6991	journal	April 1988
Deep Potential: A General Representation of a Many-Body Potential Energy Surface Han, Jiequn; Zhang, Linfeng; Car, Roberto Communications in Computational Physics, Vol. 23, Issue 3 https://doi.org/10.4208/cicp.OA-2017-0213	journal	January 2018
Machine learning of accurate energy-conserving molecular force fields Chmiela, Stefan; Tkatchenko, Alexandre; Sauceda, Huziel E. Science Advances, Vol. 3, Issue 5 https://doi.org/10.1126/sciadv.1603015	journal	May 2017
Benchmarking exchange-correlation functionals for hydrogen at high pressures using quantum Monte Carlo Clay, Raymond C.; Mcminis, Jeremy; McMahon, Jeffrey M. Physical Review B, Vol. 89, Issue 18 https://doi.org/10.1103/PhysRevB.89.184106	journal	May 2014
Computing forces with quantum Monte Carlo Assaraf, Roland; Caffarel, Michel The Journal of Chemical Physics, Vol. 113, Issue 10 https://doi.org/10.1063/1.1286598	journal	September 2000
Accurate, Efficient, and Simple Forces Computed with Quantum Monte Carlo Methods Chiesa, Simone; Ceperley, D. M.; Zhang, Shiwei Physical Review Letters, Vol. 94, Issue 3 https://doi.org/10.1103/PhysRevLett.94.036404	journal	January 2005
Stable Solid Molecular Hydrogen above 900 K from a Machine-Learned Potential Trained with Diffusion Quantum Monte Carlo Niu, Hongwei; Yang, Yubo; Jensen, Scott Physical Review Letters, Vol. 130, Issue 7 https://doi.org/10.1103/PhysRevLett.130.076102	journal	February 2023

Similar Records

Machine learning for continuous quantum error correction on superconducting qubits

Journal Article · Wed Jun 15 00:00:00 EDT 2022 · New Journal of Physics · OSTI ID:2323949

Convy, Ian; Liao, Haoran; Zhang, Song; +5 more

Error-Induced Beam Degradation in Fermilab's Accelerators

Thesis/Dissertation · Tue Jan 01 00:00:00 EST 2008 · OSTI ID:2323949

Yoon, Sung-Young Phil

Quantum-assisted associative adversarial network: applying quantum annealing in deep learning

Journal Article · Mon Jun 07 00:00:00 EDT 2021 · Quantum Machine Intelligence · OSTI ID:2323949

Wilson, Max; Vandal, Thomas; Hogg, Tad; +1 more

Title: Training models using forces computed by stochastic electronic structure methods

Citation Formats

References (40)

Similar Records

Related Subjects