Self-consistency error correction for accurate machine learning potentials from variational Monte Carlo
arXiv:2504.20481 · doi:10.1021/acs.jctc.5c00715
Abstract
Variational Monte Carlo (VMC) can be used to train accurate machine learning interatomic potentials (MLIPs), enabling molecular dynamics (MD) simulations of complex materials on time scales and for system sizes previously unattainable. VMC training sets are often based on partially optimized wave functions (WFs) to circumvent expensive energy optimizations of the whole set of WF parameters. However, frozen variational parameters lead to VMC forces and pressures not consistent with the underlying potential energy surface, a bias called the self-consistency error (SCE). Here, we demonstrate how the SCE can spoil the accuracy of MLIPs trained on these data, taking high-pressure hydrogen as test case. We then apply a recently introduced SCE correction [ Phys. Rev. B 109, 205151 (2024)] to generate unbiased VMC training sets based on a Jastrow-correlated single determinant WF with frozen Kohn-Sham orbitals. The MLIPs generated within this framework are significantly improved and can approach in quality those trained on datasets built with fully optimized WFs. Our conclusions are further supported by MD simulations, which show how MLIPs trained on SCE-corrected datasets systematically yield more reliable physical observables. Our framework opens the possibility of constructing extended high-quality training sets with VMC.
References in corpus (31)
- Quantum ESPRESSO: a modular and open-source software project for quantum simulations of materials
- Advanced capabilities for materials modelling with Quantum ESPRESSO
- Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
- On representing chemical environments
- Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
- Quantum ESPRESSO toward the exascale
- E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
- Machine Learning of Accurate Energy-Conserving Molecular Force Fields
- Machine learning for molecular simulation
- Big Data meets Quantum Chemistry Approximations: The -Machine Learning Approach
- Comparing molecules and solids across structural and alchemical space
- FPEOS: A First-Principles Equation of State Table of Deuterium for Inertial Confinement Fusion Applications
- Stable liquid Hydrogen at high pressure by a novel ab-initio molecular dynamics
- Zero-Variance Zero-Bias Principle for Observables in quantum Monte Carlo: Application to Forces
- Backflow Correlations for the Electron Gas and Metallic Hydrogen
- Algorithmic differentiation and the calculation of forces by quantum Monte Carlo
- First principles simulations of dense hydrogen
- Ultra-fast interpretable machine-learning potentials
- Data-driven Material Models for Atomistic Simulation
- TurboRVB: a many-body toolkit for {\it ab initio} electronic simulations by quantum Monte Carlo
- Quantum phase diagram of high-pressure hydrogen
- Simple formalism for efficient derivatives and multi-determinant expansions in quantum Monte Carlo
- Systematically convergent method for accurate total energy calculations with localized atomic orbitals
- Atomic forces by quantum Monte Carlo: application to phonon dispersion calculation
- Towards Quantum Monte Carlo Forces on Heavier Ions: Scaling Properties
- Geminal embedding scheme for optimal atomic basis set construction in correlated calculations
- Space-warp coordinate transformation for efficient ionic force calculations in quantum Monte Carlo
- Accurate quantum Monte Carlo forces for machine-learned force fields: Ethanol as a benchmark
- Principal deuterium Hugoniot via Quantum Monte Carlo and -learning
- Efficient calculation of unbiased atomic forces in ab initio Variational Monte Carlo
- Interpolating many-body wave functions for accelerated molecular dynamics on the near-exact electronic surface