Predicting reaction conditions from limited data through active transfer learning. Issue 22 (23rd May 2022)

• Record Type: Journal Article
• Authors: Shim, Eunjae; Kammeraad, Joshua A.; Xu, Ziping; Tewari, Ambuj; Cernak, Tim; Zimmerman, Paul M.
• Is Part Of: Chemical science. Volume 13:Issue 22(2022)
• Page Start: 6655
• Page End: 6668
• Publication Date: 2022-05-23
• Subjects: Chemistry -- Periodicals; 540.5
• DOI: 10.1039/d1sc06932b
• Languages: English
• ISSNs: 2041-6520

Abstract

Abstract : Transfer learning is combined with active learning to discover synthetic reaction conditions in a small-data regime. This strategy is tested on cross-coupling reactions from a high-throughput experimentation dataset and shows promising results. Abstract : Transfer and active learning have the potential to accelerate the development of new chemical reactions, using prior data and new experiments to inform models that adapt to the target area of interest. This article shows how specifically tuned machine learning models, based on random forest classifiers, can expand the applicability of Pd-catalyzed cross-coupling reactions to types of nucleophiles unknown to the model. First, model transfer is shown to be effective when reaction mechanisms and substrates are closely related, even when models are trained on relatively small numbers of data points. Then, a model simplification scheme is tested and found to provide comparative predictivity on reactions of new nucleophiles that include unseen reagent combinations. Lastly, for a challenging target where model transfer only provides a modest benefit over random selection, an active transfer learning strategy is introduced to improve model predictions. Simple models, composed of a small number of decision trees with limited depths, are crucial for securing generalizability, interpretability, and performance of active transfer learning.