Fig. 3. Evolution of the workflow of machine learning in nanomaterials discovery and design. (a) First-generation approach. In this paradigm, there are two main steps: feature engineering from raw database to descriptors; model building from descriptors to target model. (b) Second-generation approach. The key characteristic that distinguishes it from the first-generation approach is eliminating human-expert feature engineering, which can be directly learned from raw nanomaterials.²⁹.