EUVIMEDEUVIMED entdecken
Crossref

Empowering Reactivity Predictions through Noise-Based Data Augmentation

Julian A Hueffel iD, Quentin P Bindschaedler iD, Francesco Sala iD, Franziska Schoenebeck iD

Journal of the American Chemical Society · 10.09.2025 · 147 (36) · 33232-33239

DOI: 10.1021/jacs.5c11632 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the American Chemical Society, online erschienen am 02.09.2025. Verfasst von Julian A Hueffel und 3 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Organic Chemistry, RWTH Aachen University, Landoltweg 1, 52074 Aachen, Germany. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 48 Literaturangaben. Gefördert von: Deutsche Forschungsgemeinschaft.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Data scarcity is a key obstacle in the pursuit to capitalize on the predictive powers of the A.I. in problems related to bond-making and -breaking at the molecular level. While the generation of artificial data from real data points (known as "data augmentation") is a widely pursued strategy employed in, for example, image or speech recognition or health data evaluations, among others, to artificially expand existing data sets, it is currently unknown whether this strategy is applicable to reactivity problems at the molecular level, where predictive models are exquisitely sensitive to steric, electronic, and structural nuances. Here, we systematically evaluated the power of data augmentation for a diverse set of reactivity questions ranging from the prediction of activation barriers to stereoselectivities of catalytic transformations. We demonstrate that introducing Gaussian noise to existing data points, which is completed in under a second for a full data set, can dramatically enhance the predictive performance. It can enable model training in low-data regimes where otherwise no meaningful model could be built and achieves accuracy comparable to models built on full data sets, while needing only a fraction of the data. The approach substantially lowers the number of necessary experiments (by 20-50%), conserving time, energy, and resources, while advancing the integration of machine learning in molecular reactivity challenges.

Autor:innen und Institutionen

  • Institute of Organic Chemistry, RWTH Aachen University, Landoltweg 1, 52074 Aachen, Germany.

ORCID-Profile: Julian A Hueffel, Quentin P Bindschaedler, Francesco Sala, Franziska Schoenebeck

Förderung

  • Deutsche Forschungsgemeinschaft

Zitieren

APA

Hueffel, J. A., Bindschaedler, Q. P., Sala, F., & Schoenebeck, F. (2025). Empowering Reactivity Predictions through Noise-Based Data Augmentation. Journal of the American Chemical Society, 147(36), 33232-33239. https://doi.org/10.1021/jacs.5c11632

Vancouver

Hueffel JA, Bindschaedler QP, Sala F, Schoenebeck F. Empowering Reactivity Predictions through Noise-Based Data Augmentation. Journal of the American Chemical Society. 2025;147(36):33232-33239. doi:10.1021/jacs.5c11632

BibTeX
@article{Hueffel2025,
  author = {Hueffel, Julian A and Bindschaedler, Quentin P and Sala, Francesco and Schoenebeck, Franziska},
  title = {Empowering Reactivity Predictions through Noise-Based Data Augmentation},
  journal = {Journal of the American Chemical Society},
  year = {2025},
  volume = {147},
  number = {36},
  pages = {33232--33239},
  doi = {10.1021/jacs.5c11632}
}
Kostenlos starten →