Bayesian machine learning for quantum molecular dynamics

Abstract

This article discusses applications of Bayesian machine learning for quantum molecular dynamics. One particular formulation of quantum dynamics advocated here is in the form of a machine learning simulator of the Schrödinger equation. If combined with the Bayesian statistics, such a simulator allows one to obtain not only the quantum predictions but also the error bars of the dynamical results associated with uncertainties of inputs (such as the potential energy surface or non-adiabatic couplings) into the nuclear Schrödinger equation. Instead of viewing atoms as undergoing dynamics on a given potential energy surface, Bayesian machine learning allows one to formulate the problem as the Schrödinger equation with a non-parametric distribution of potential energy surfaces that becomes conditioned by the desired dynamical properties (such as the experimental measurements). Machine learning models of the Schrödinger equation solutions can identify the sensitivity of the dynamical properties to different parts of the potential surface, the collision energy, angular momentum, external field parameters and basis sets used for the calculations. This can be used to inform the design of efficient quantum dynamics calculations. Machine learning models can also be used to correlate rigorous results with approximate calculations, providing accurate interpolation of exact results. Finally, there is evidence that it is possible to build Bayesian machine learning models capable of physically extrapolating the solutions of the Schrödinger equation. This is particularly valuable as such models could complement common discovery tools to explore physical properties at Hamiltonian parameters not accessible by rigorous quantum calculations or experiments, and potentially be used to accelerate the numerical integration of the nuclear Schrödinger equation.

45 pages, 8 figures

References in corpus (24)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Kernel methods in machine learning
• Big Data meets Quantum Chemistry Approximations: The $\Delta$-Machine Learning Approach
• Machine Learning of Molecular Electronic Properties in Chemical Compound Space
• Machine Learning Energies of 2 M Elpasolite (ABC${}_2$D${}_6$) Crystals
• Neural networks for the prediction organic chemistry reactions
• Bayesian Error Estimation in Density Functional Theory
• Structure-based Sampling and Self-correcting Machine Learning for Accurate Calculations of Potential Energy Surfaces and Vibrational Levels
• Efficient non-parametric fitting of potential energy surfaces for polyatomic molecules with Gaussian processes
• Production of ultracold NH molecules by sympathetic cooling with Mg
• Calculating vibrational spectra with sum of product basis functions without storing full-dimensional vectors or matrices
• Extrapolating quantum observables with machine learning: Inferring multiple phase transitions from properties of a single phase
• Reliable estimation of prediction uncertainty for physico-chemical property models
• Bayesian optimization for the inverse scattering problem in quantum reaction dynamics
• Gaussian Process Model for Collision Dynamics of Complex Molecules
• Quantum theory of chemical reactions in the presence of electromagnetic fields
• The prospects of sympathetic cooling of NH molecules with Li atoms
• The parameters uncertainty inflation fallacy
• Tuning bimolecular chemical reactions by electric fields
• Cold collisions of N atoms and NH molecules in magnetic fields
• Optimization of an exchange-correlation density functional for water
• Cold collisions of complex polyatomic molecules
• Elastic and inelastic collisions of ${}^2\Sigma$ molecules in a magnetic field

Cited by in corpus (26)

• Machine learning for electronically excited states of molecules
• Bayesian optimization for tuning and selecting hybrid-density functionals
• Combining SchNet and SHARC: The SchNarc machine learning approach for excited-state dynamics
• Quantum Machine Learning for Chemistry and Physics
• Neural Network Potentials for Chemistry: Concepts, Applications and Prospects
• Fusion of Probability Density Functions
• Neural networks and kernel ridge regression for excited states dynamics of CH${}_2$NH${}_2^{+}$: From single-state to multi-state representations and multi-property machine learning models
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Machine learning and excited-state molecular dynamics
• Accelerating Atomistic Simulations with Piecewise Machine Learned Ab Initio Potentials at Classical Force Field-like Cost
• Gaussian process model of 51-dimensional potential energy surface for protonated imidazole dimer
• Machine-learning-corrected quantum dynamics calculations
• A data-driven approach to determine dipole moments of diatomic molecules
• Bayesian optimization for inverse problems in time-dependent quantum dynamics
• Hessian-based toolbox for reliable and interpretable machine learning in physics
• Accelerated Bayesian Inference for Molecular Simulations using Local Gaussian Process Surrogate Models
• Quantum Extreme Learning of molecular potential energy surfaces and force fields
• Gradient domain machine learning with composite kernels: improving the accuracy of PES and force fields for large molecules
• Transfer matrix theory of surface spin echo experiments with molecules
• Leveraging Legacy Data to Accelerate Materials Design via Preference Learning
• Rainbow scattering in rotationally inelastic collisions of HCl and H${}_2$
• Quantum transfer through small networks coupled to phonons: effects of topology vs phonons
• Echoes in a parametrically perturbed Kerr-nonlinear oscillator
• Multi-objective optimization for retinal photoisomerization models with respect to experimental observables
• Quantitative Molecular Simulations
• Neural network ensemble for computing cross sections for rotational transitions in H${}_2$O + H${}_2$O collisions