TSNet: predicting transition state structures with tensor field networks and transfer learning

Jackson, Riley; Zhang, Wenyuan; Pearson, Jason K.

doi:10.1039/d1sc01206a

Cited by 39 publications

(53 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also would like to note that very recent work has used ML to predict transition state structures directly. 49,50 Should these models become accurate on a broader scale they could provide an alternative path towards predicting TST rate constants.…”

Section: Discussionmentioning

confidence: 99%

Low-cost prediction of molecular and transition state partition functions via machine learning

Komp

Valleau

2022

Chem. Sci.

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Low-cost prediction of molecular and transition state partition functions via machine learning

Komp

Valleau

2022

Chem. Sci.

View full text Add to dashboard Cite

show abstract

“…To compare the performances against the other ML models, two existing models, TSGen and TSNet, are trained with the same train dataset. [24,25]. TSGen exhibited slightly better performance than TSNet.…”

Section: Prediction Accuracymentioning

confidence: 95%

“…This model, referred to as TSGen hereinafter, adopt internal nonlinear optimization to fine the atomic positions with the closet interatomic distance matrix to the initial one. A second model, TSNet was released by Jackson et al [25] This model is based on a tensor-field network that applies spherical harmonics as convolution filters to distinguish relative atomic positions and directly predicts the atomic positions of TS structures. [26] These ML architectures mathematically preserve the conditions for determining TS structures.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Transition State Structures of General Chemical Reactions via Machine Learning

Choi

2022

Preprint

View full text Add to dashboard Cite

The elucidation of transition state (TS) structures is ssential for understanding the mechanisms of chemical reactions and exploring reaction networks. Despite advances in computational approaches, TS searches remain still a challenging problem due to the difficulty of constructing an initial structure and heavy computational costs. Herein, a novel machine learning (ML) model for predicting TS structures of general organic reactions is proposed. The proposed model derives interatomic distances of a TS structure from atomic pair features reflecting reactant, product, and linearly interpolated structures. The model shows excellent accuracy, particularly for the atomic pairs where bond formation or breakage occurs. The predicted TS structures result in a high success ratio (93.8%) of quantum chemical saddle-point optimizations, and 88.8% of the optimization results have energy errors of less than 0.1 kcal/mol. Additionally, as a proof-of-concept, exploring multiple reaction paths of an organic reaction is demonstrated with the ML inferences. I envision that the proposed approach will aid the construction of initial geometry for TS optimization and reaction path explorations.

show abstract

“…Cusworth and Dodsworth examined typical constructions and atypical "hyphenated" sentences, thereby thinking about the research scope of construction grammar, the general mechanism of the itinerary, and the cross-linguistic characteristics of constructions [11]. With the further development of this theory, scholars have gradually expanded its scope of application to case studies of specific structures, such as a series of studies by Jackson et al, the constructional meaning of each component in unique sentence structures such as transitive construction, "ba," and "bei" is discussed [12]. e relationship between the components and the constructional meaning is analyzed.…”

Section: Related Workmentioning

confidence: 99%

Exploring the Application of the Random Matrix Thinking Model in Teaching English Predicate Constructions

Liu

Wen

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

This paper, the random matrix thinking model, studies the grammatical features of verbs on random matrix and logical relations and designs a model for teaching English predicate constructions by combining random matrix thinking and arithmetic rules. In this paper, an attention mechanism-based English predicate verb recognition method and a BERT-based English predicate verb recognition method are proposed for the characteristics of predicate verbs as the focal point of sentences. Since the attention mechanism can obtain the long-distance semantic dependency information in the sentence, the attention mechanism-based method can effectively improve the recognition performance of the predicate verb compared with the traditional way. The BERT-based English predicative verb recognition model improves the former approach by taking full advantage of the input corpus and improving the model performance. A random matrix-based predicate verb uniqueness discrimination method is proposed for the uniqueness of predicate verbs in sentences. By setting the classification fitting conditions, the data can be optimized to output and fully fit the global identity of the predicate verb in the penalty during the training process, and better results are achieved.

show abstract

TSNet: predicting transition state structures with tensor field networks and transfer learning

Abstract: Transition states are among the most important molecular structures in chemistry, critical to a variety of fields such as reaction kinetics, catalyst design, and the study of protein function. However,...

Cited by 39 publications

References 67 publications

Low-cost prediction of molecular and transition state partition functions via machine learning

Low-cost prediction of molecular and transition state partition functions via machine learning

Prediction of Transition State Structures of General Chemical Reactions via Machine Learning

Exploring the Application of the Random Matrix Thinking Model in Teaching English Predicate Constructions

Contact Info

Product

Resources

About