Barrier height prediction by machine learning correction of semiempirical calculations

García-Andrade, Xabier; Tahoces, Pablo G.; Pérez‐Ríos, Jesús; Martı́nez-Núñez, Emilio

doi:10.48550/arxiv.2208.02289

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, other groups have analyzed performance on geometries from semi-empirical methods or even force fields [54], which are much more practical for users to obtain than ab initio geometries. Other studies have used a ∆ML technique [55], such as using semi-empirical geometries to predict DFT quality energies [56][57][58] or barrier heights [59,60], bringing a potential energy surface from B3LYP up to CCSD(T) [61], or predicting the effect of perturbatively included triples [62]. Regardless, we want to conclude this section by stressing that any model operating on 3D coordinates will always require an extra step compared to models operating on simpler representations.…”

Section: Reconsidering Model Inputsmentioning

confidence: 99%

Comment on ‘Physics-based representations for machine learning properties of chemical reactions’

Spiekermann,

Stuyver,

Pattanaik

et al. 2023

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

In a recent article in this journal, van Gerwen et al (2022 Mach. Learn.: Sci. Technol. 3 045005) presented a kernel ridge regression model to predict reaction barrier heights. Here, we comment on the utility of that model and present references and results that contradict several statements made in that article. Our primary interest is to offer a broader perspective by presenting three aspects that are essential for researchers to consider when creating models for chemical kinetics: (1) are the model’s prediction targets and associated errors sufficient for practical applications? (2) Does the model prioritize user-friendly inputs so it is practical for others to integrate into prediction workflows? (3) Does the analysis report performance on both interpolative and more challenging extrapolative data splits so users have a realistic idea of the likely errors in the model’s predictions?

show abstract