Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights

Sauceda, Huziel E.; Chmiela, Stefan; Poltavsky, Igor; Müller, Klaus‐Robert; Tkatchenko, Alexandre

doi:10.1007/978-3-030-40245-7_14

Cited by 20 publications

(25 citation statements)

References 92 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Today, a couple hundred ab initio reference calculations are enough to construct ML-FFs that reach this accuracy within a few tens of wavenumbers. 327 In the past, even if suitable reference data was available, constructing accurate force fields was labor-intensive and required human effort and expertise. Nowadays, by virtue of automatic ML methods, the same task is as effortless as the push of a button.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning Force Fields

et al. 2021

Self Cite

View full text Add to dashboard Cite

In recent years, the use of machine learning (ML) in computational chemistry has enabled numerous advances previously out of reach due to the computational complexity of traditional electronic-structure methods. One of the most promising applications is the construction of ML-based force fields (FFs), with the aim to narrow the gap between the accuracy of ab initio methods and the efficiency of classical FFs. The key idea is to learn the statistical relation between chemical structure and potential energy without relying on a preconceived notion of fixed chemical bonds or knowledge about the relevant interactions. Such universal ML approximations are in principle only limited by the quality and quantity of the reference data used to train them. This review gives an overview of applications of ML-FFs and the chemical insights that can be obtained from them. The core concepts underlying ML-FFs are described in detail, and a step-by-step guide for constructing and testing them from scratch is given. The text concludes with a discussion of the challenges that remain to be overcome by the next generation of ML-FFs.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Learning Force Fields

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…The coupled cluster datasets were generated by the representative sampling method as reported in ref. [21].…”

Section: Methodology a Machine Learned Force Fields: Sgdmlmentioning

confidence: 99%

“…coupled cluster) which is not always computationally affordable when performing long ab initio path integral molecular dynamics (PIMD) simulations. In this study, we have performed PIMD simulations using machine learned molecular force fields constructed using the sGDML framework [17][18][19][20][21][22][23][24] and trained on coupled cluster reference data [CCSD(T) or CCSD depending on the size of the molecule] [25,26](see Supporting Information for more details).…”

Section: Introductionmentioning

confidence: 99%

Dynamical Strengthening of Covalent and Non-Covalent Molecular Interactions by Nuclear Quantum Effects at Finite Temperature

Sauceda,

Vassilev-Galindo,

Chmiela

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Nuclear quantum effects (NQE) tend to generate delocalized molecular dynamics due to the inclusion of the zero point energy and its coupling with the anharmonicities in interatomic interactions. Here, we present evidence that NQE often enhance electronic interactions and, in turn, can result in dynamical molecular stabilization at finite temperature. The underlying physical mechanism promoted by NQE depends on the particular interaction under consideration. First, the effective reduction of interatomic distances between functional groups within a molecule can enhance the n → π * interaction by increasing the overlap between molecular orbitals or by strengthening electrostatic interactions between neighboring charge densities. Second, NQE can localize methyl rotors by temporarily changing molecular bond orders and leading to the emergence of localized transient rotor states. Third, for noncovalent van der Waals interactions the strengthening comes from the increase of the polarizability given the expanded average interatomic distances induced by NQE. The implications of these boosted interactions include counterintuitive hydroxyl-hydroxyl bonding, hindered methyl rotor dynamics, and molecular stiffening which generates smoother free-energy surfaces. Our findings yield new insights into the versatile role of nuclear quantum fluctuations in molecules and materials.

show abstract

“…Recently, we have systematically demonstrated that sGDML models trained on only few 100s of reference structures are able to reconstruct molecular PESs with a mean absolute error (MAE) of less that 0.06 kcal mol −1 for small molecules with up to 15 atoms and less than 0.16 kcal mol −1 for molecules as complex as aspirin, paracetamol, and azobenzene. 9,32,64,69,70 These results will be of particular importance when directly comparing with less flexible MM-FFs in Sec. IV E.…”

Section: Article Scitationorg/journal/jcpmentioning

confidence: 96%

Molecular force fields with gradient-domain machine learning (GDML): Comparison and synergies with classical force fields

Sauceda

Gastegger

Chmiela

et al. 2020

The Journal of Chemical Physics

Self Cite

View full text Add to dashboard Cite

Modern machine learning force fields (ML-FF) are able to yield energy and force predictions at the accuracy of high-level ab initio methods, but at a much lower computational cost. On the other hand, classical molecular mechanics force fields (MM-FF) employ fixed functional forms and tend to be less accurate, but considerably faster and transferable between molecules of the same class. In this work, we investigate how both approaches can complement each other. We contrast the ability of ML-FF for reconstructing dynamic and thermodynamic observables to MM-FFs in order to gain a qualitative understanding of the differences between the two approaches. This analysis enables us to modify the generalized AMBER force field by reparametrizing short-range and bonded interactions with more expressive terms to make them more accurate, without sacrificing the key properties that make MM-FFs so successful.

show abstract

Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights

Cited by 20 publications

References 92 publications

Machine Learning Force Fields

Machine Learning Force Fields

Dynamical Strengthening of Covalent and Non-Covalent Molecular Interactions by Nuclear Quantum Effects at Finite Temperature

Molecular force fields with gradient-domain machine learning (GDML): Comparison and synergies with classical force fields

Contact Info

Product

Resources

About