General Chemically Intuitive Atom- and Bond-Level DFT Descriptors for Machine Learning Approaches to Reaction Condition Prediction

We demonstrate the usefulness of general atom- and bond-level density functional theory (DFT) descriptors to enhance the performance of neural networks for general reaction condition prediction. We treat condition prediction as a multiclass classification task and report the performance of neural ne...

Full description

Saved in:

Bibliographic Details
Main Author:	Miguel Nouman (21557202) (author)
Other Authors:	Richard B. Canty (19337513) (author), Brent A. Koscher (1449973) (author), Matthew A. McDonald (10766331) (author), Klavs F. Jensen (722008) (author)
Published:	2025
Subjects:	Genetics Biotechnology Biological Sciences not elsewhere classified Information Systems not elsewhere classified unsupervised learning algorithms treat condition prediction share considerable conceptual reaction condition prediction multiclass classification task machine learning approaches fewer trainable parameter 296 distinct single hybrid embeddings outperforms purely structural counterparts combine general dft superior weighted precision neural networks trained level dft descriptors general dft descriptors embedding hybrid descriptors 1 </ sub weighted precision neural networks hybrid representations general atom structural descriptors f </ embedding strategy >< sub combining structural random forests provide comparable one used latter benefiting input sizes fold cross equivalent architectures data points architectural similarities 3 accuracies 267 times 1 accuracy
Tags:	Add Tag No Tags, Be the first to tag this record!

General Chemically Intuitive Atom- and Bond-Level DFT Descriptors for Machine Learning Approaches to Reaction Condition Prediction

Similar Items