Automatic Feature Engineering for Catalyst Design Using Small Data without Prior Knowledge of the Target Catalysis

The empirical aspect of descriptor design with limited data in catalyst informatics entails a logical contradiction as it relies on sufficient prior knowledge for exploring the unknown. In this study, we developed a technique for automatic feature engineering (AFE) that works on small catalyst data without requiring any prior knowledge of the target catalysis. This technique generates a large number of features through mathematical operations on general physicochemical fea-tures of catalytic components, and extracts the relevant features for the desired catalysis, essentially screening a large number of hypotheses on a machine. AFE yielded reasonable regression results for three types of heterogeneous cataly-sis: oxidative coupling of methane (OCM), conversion of ethanol to butadiene, and three-way catalysis, where only the training set was swapped. Moreover, through the application of active learning that combines AFE and high-throughput experimentation for OCM, we successfully visualized the machine’s process of acquiring precise recognition of catalyst design. AFE is a versatile technique for data-driven catalysis research and a key step towards fully automated catalyst discoveries.