∆-Machine Learning for the Prediction of Metal Complex Properties
Abstract
∆-machine learning was applied to predict electronic properties of TMCs, improving accuracy and chemical transferability compared to traditional models. Additionally, it was demonstrated how minimizing the cost of the low-fidelity method substantially improves computational efficiency, while only slightly affecting predictive performance.
Fields of Study & Subjects
Machine Learning in Materials Science
Computational Drug Discovery Methods
Advanced Memory and Neural Computing
Publication details
- Journal / source
- ChemRxiv
- Publisher
- American Chemical Society (ACS)
- Publication date
- 28 January 2026
- Document type
- Preprint