Representing globally accurate reactive potential energy surfaces with complex topography by combining Gaussian process regression and neural networks

Yang, Zijiang; Chen, Hanghang; Chen, Maodu

doi:10.1039/d2cp00719c

Cited by 7 publications

(5 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, there has been increasing attention on representing molecular PESs by machine learning technology. ,− Among those algorithms, the artificial NN model features high fitting accuracy and strong generalization performance and has become the most popular method in constructing reactive PESs of small systems. In this work, the back-propagation NN method is used to map the global ground-state CaH 2 PES.…”

Section: Theoretical Methods and Computational Detailsmentioning

confidence: 99%

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(¹S) + H₂(v₀ = 0, j₀ = 0) → CaH + H Reaction

Chen,

Mao,

Yang

et al. 2024

ACS Omega

Self Cite

View full text Add to dashboard Cite

The reactive collision between Ca and H 2 molecules has attracted great interest experimentally due to the key role of the product CaH molecule in the field of astrophysics and cold molecules. However, quantum dynamics calculations for this system have not been reported due to the lack of a global potential energy surface (PES). Herein, a globally accurate PES of the ground-state CaH 2 is developed by combining 11365 high-level ab initio points and permutation invariant polynomial neural network method. Based on the newly constructed PES, the state-to-state quantum dynamics calculations for the Ca( 1 S) + H 2 (v 0 = 0, j 0 = 0) → CaH + H reaction are carried out using the time-dependent wave packet method. The dynamic results reveal that the reaction follows the complex-forming mechanism near the reactive threshold, whereas both the indirect insertion mechanism and direct abstraction mechanism have effects at higher collision energies. The newly constructed PES can be used to further study the influence of isotope substitution, rovibrational excitation, and spatial orientation of reactant molecules on reaction dynamics.

show abstract

Section: Theoretical Methods and Computational Detailsmentioning

confidence: 99%

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(¹S) + H₂(v₀ = 0, j₀ = 0) → CaH + H Reaction

Chen,

Mao,

Yang

et al. 2024

ACS Omega

Self Cite

View full text Add to dashboard Cite

show abstract

“…The main challenge in constructing multi-dimensional PESs is to represent the function between the potential energies and the molecular nuclear coordinates based on the discrete ab initio data. Fitting PESs with machine learning models has been gaining popularity in recent years, and using an artificial neural network (NN) [ 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 ] or a Gaussian process (GP) [ 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 ] are the two most common approaches. GP is a kernel-based supervised statistical learning method [ 71 ], which has been widely used to solve physical chemistry problems such as mapping high-dimensional PESs and simulating quantum scattering dynamics.…”

Section: Ground-state Lina 2 Pesmentioning

confidence: 99%

Globally Accurate Gaussian Process Potential Energy Surface and Quantum Dynamics Studies on the Li(2S) + Na2 → LiNa + Na Reaction at Low Collision Energies

et al. 2023

Self Cite

View full text Add to dashboard Cite

The LiNa2 reactive system has recently received great attention in the experimental study of ultracold chemical reactions, but the corresponding theoretical calculations have not been carried out. Here, we report the first globally accurate ground-state LiNa2 potential energy surface (PES) using a Gaussian process model based on only 1776 actively selected high-level ab initio training points. The constructed PES had high precision and strong generalization capability. On the new PES, the quantum dynamics calculations on the Li(2S) + Na2(v = 0, j = 0) → LiNa + Na reaction were carried out in the 0.001–0.01 eV collision energy range using an improved time-dependent wave packet method. The calculated results indicate that this reaction is dominated by a complex-forming mechanism at low collision energies. The presented dynamics data provide guidance for experimental research, and the newly constructed PES could be further used for ultracold reaction dynamics calculations on this reactive system.

show abstract

“…The primary utility of ML in this regard is the flexibility of neural network (NN) [42] constructions in their ability to accurately describe complex functional relationships [43][44][45][46]. NN PES constructions have been demonstrated to achieve required accuracy for a range of relevant systems [26,[47][48][49][50], including in particular reactive PES studies [26,[48][49][50].…”

Section: Introductionmentioning

confidence: 99%

Multifidelity neural network formulations for prediction of reactive molecular potential energy surfaces

Yang

Eldred

Zádor

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper focuses on the development of multifidelity modeling approaches using neural network surrogates, where training data arising from multiple model forms and resolutions are integrated to predict high-fidelity response quantities of interest at lower cost. We focus on the context of quantum chemistry and the integration of information from multiple levels of theory. Important foundations include the use of symmetry function-based atomic energy vector constructions as feature vectors for representing structures across families of molecules, and single-fidelity neural network training capabilities that learn the relationships needed to map feature vectors to potential energy predictions. These foundations are embedded within several multifidelity topologies that decompose the high-fidelity mapping into model-based components, including sequential formulations that admit a general nonlinear mapping across fidelities and discrepancy-based formulations that presume an additive decomposition. Methodologies are first explored and demonstrated on a pair of simple analytical test problems, and then deployed for potential energy prediction for C5H5 using B2PLYPD3 6-311++G(d,p) for high fidelity simulation data and Hartree Fock 6-31G for low fidelity data. For the common case of limited access to high-fidelity data, our computational results demonstrate that multifidelity neural network potential energy surface constructions achieve roughly an order of magnitude improvement, either in terms of test error reduction for equivalent total simulation cost or reduction in total cost for equivalent error.

show abstract

Representing globally accurate reactive potential energy surfaces with complex topography by combining Gaussian process regression and neural networks

Abstract: There has been increasing attention in using machine learning technologies, such as neural network (NN) and Gaussian process regression (GPR), to model multidimensional potential energy surfaces (PESs). NN PES features...

Cited by 7 publications

References 73 publications

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(¹S) + H₂(v₀ = 0, j₀ = 0) → CaH + H Reaction

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(¹S) + H₂(v₀ = 0, j₀ = 0) → CaH + H Reaction

Globally Accurate Gaussian Process Potential Energy Surface and Quantum Dynamics Studies on the Li(2S) + Na2 → LiNa + Na Reaction at Low Collision Energies

Multifidelity neural network formulations for prediction of reactive molecular potential energy surfaces

Contact Info

Product

Resources

About

Representing globally accurate reactive potential energy surfaces with complex topography by combining Gaussian process regression and neural networks

Abstract: There has been increasing attention in using machine learning technologies, such as neural network (NN) and Gaussian process regression (GPR), to model multidimensional potential energy surfaces (PESs). NN PES features...

Cited by 7 publications

References 73 publications

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(1S) + H2(v0 = 0, j0 = 0) → CaH + H Reaction

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(1S) + H2(v0 = 0, j0 = 0) → CaH + H Reaction

Globally Accurate Gaussian Process Potential Energy Surface and Quantum Dynamics Studies on the Li(2S) + Na2 → LiNa + Na Reaction at Low Collision Energies

Multifidelity neural network formulations for prediction of reactive molecular potential energy surfaces

Contact Info

Product

Resources

About

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(¹S) + H₂(v₀ = 0, j₀ = 0) → CaH + H Reaction

A Neural Network Potential Energy Surface and Quantum Dynamics Study of Ca(¹S) + H₂(v₀ = 0, j₀ = 0) → CaH + H Reaction