Efficient non-parametric fitting of potential energy surfaces for polyatomic molecules with Gaussian processes

Abstract

We explore the performance of a statistical learning technique based on Gaussian Process (GP) regression as an efficient non-parametric method for constructing multi-dimensional potential energy surfaces (PES) for polyatomic molecules. Using an example of the molecule N${}_4$, we show that a realistic GP model of the six-dimensional PES can be constructed with only 240 potential energy points. We construct a series of the GP models and illustrate the convergence of the accuracy of the resulting surfaces as a function of the number of $abinitio$ points. We show that the GP model based on $\sim 1500$ potential energy points achieves the same level of accuracy as the conventional regression fits based on 16,421 points. The GP model of the PES requires no fitting of $abinitio$ data with analytical functions and can be readily extended to surfaces of higher dimensions.

arXiv comment2 figures

References in corpus (3)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Gaussian Process Model for Collision Dynamics of Complex Molecules
• Gaussian Process Model for Extrapolation of Scattering Observables for Complex Molecules: from Benzene to Benzonitrile

Cited by in corpus (34)

• Machine learning for electronically excited states of molecules
• Embedded Atom Neural Network Potentials: Efficient and Accurate Machine Learning with a Physically Inspired Representation
• Structure-based Sampling and Self-correcting Machine Learning for Accurate Calculations of Potential Energy Surfaces and Vibrational Levels
• Neural Network Potentials for Chemistry: Concepts, Applications and Prospects
• Operators in Machine Learning: Response Properties in Chemical Space
• Bayesian machine learning for quantum molecular dynamics
• Automatically growing global reactive neural network potential energy surfaces: a trajectory free active learning strategy
• Building machine learning force fields for nanoclusters
• Gaussian process model of 51-dimensional potential energy surface for protonated imidazole dimer
• PhysNet Meets CHARMM: A Framework for Routine Machine Learning / Molecular Mechanics Simulations
• Barrier height prediction by machine learning correction of semiempirical calculations
• Invariant Manifolds and Rate Constants in Driven Chemical Reactions
• Bayesian optimization for inverse problems in time-dependent quantum dynamics
• GPU-Accelerated Approximate Kernel Method for Quantum Machine Learning
• Neural network approach to time-dependent dividing surfaces in classical reaction dynamics
• Machine learning optimization of the collocation point set for solving the Kohn-Sham equation
• Four-body singlet potential energy surface for reactions of calcium monofluoride
• Enabling rapid and accurate construction of CCSD(T)-level potential energy surface of large molecules using molecular tailoring approach
• Quantum Gaussian process model of potential energy surface for a polyatomic molecule
• The loss of the property of locality of the kernel in high-dimensional Gaussian process regression on the example of the fitting of molecular potential energy surfaces
• Quantum Extreme Learning of molecular potential energy surfaces and force fields
• Neural network approach for the dynamics on the normally hyperbolic invariant manifold of periodically driven systems
• A Machine Learning Approach for Rate Constants III: Application to the Cl(${}^2$P) + CH${}_4$ $\to$ CH${}_3$ + HCl Reaction
• Gaussian Process Regression Adaptive Density-Guided Approach: Towards Calculations of Potential Energy Surfaces for Larger Molecules
• Gradient domain machine learning with composite kernels: improving the accuracy of PES and force fields for large molecules
• A machine-learned kinetic energy model for light weight metals and compounds of group III-V elements
• Molecular dynamics-driven global tetra-atomic potential energy surfaces: Application to the AlF dimer
• Sticking lifetime of ultracold CaF molecules in triplet interactions
• MOLPIPx: an end-to-end differentiable package for permutationally invariant polynomials in Python and Rust
• Gaussian Processes for Finite Size Extrapolation of Many-Body Simulations
• Efficient interpolation of molecular properties across chemical compound space with low-dimensional descriptors
• REANN: A PyTorch-based End-to-End Multi-functional Deep Neural Network Package for Molecular, Reactive and Periodic Systems
• Rate Constants for Fine-Structure Excitations in O-H Collisions with Error Bars Obtained by Machine Learning
• Active learning of potential-energy surfaces of weakly-bound complexes with regression-tree ensembles