The impact of domain-driven and data-driven feature selection on the inverse design of nanoparticle catalysts

• Published in: Journal of computational science Vol. 65; p. 101896
• Main Authors: Li, Sichao, Ting, Jonathan Y.C., Barnard, Amanda S.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.11.2022
• Subjects: Catalysis; Inverse design; Machine learning; Nanoparticle; Platinum; Catalysis; Inverse design; Platinum; Nanoparticle; Machine learning
• ISSN: 1877-7503; 1877-7511
• DOI: 10.1016/j.jocs.2022.101896

Incorporating practical considerations into machine learning can make predictions more actionable. However, researcher interventions in the learning process may have negative impacts on model performance, leading to a trade-off between accuracy and utility. In this paper we use multi-target machine learning to predict the structure of platinum nanocatalysts based on property indicators and develop intervention scenarios using ratios of data-driven (optimal) and domain-driven (preferable) variables during feature selection. We show that minor interventions to data-driven feature selection can be tolerated, and even improve model performance, but aggressive domain-driven feature selection degrades performance, even if the mapping function is perfectly balanced. [Display omitted] •Multi-target prediction of catalysis properties with data-driven feature selection.•Inversely predicting multiple physicochemical features using subsets of properties.•Quantified impact of domain-driven feature selection on the predictive models.