Inverse Design of Nanoparticles Using Charge Transfer Properties and Multi-Target Machine Learning

Abstract

We present a new approach to inverse design that uses sets of properties to predict a unique nanoparticle structure, and is based on established multi-target regression and reliable forward structure/property prediction. Feature selection is used to focus the model on the most important characteristics before inverting the problem and simultaneously predicting multiple structural features of a single nanoparticle. The workflow is general, as demonstrated on two nanoparticle data sets, and can rapidly predict property/structure relationships to guide further research and development without the need for additional optimisation or high-throughput sampling.