Accelerating materials property predictions using machine learning

TLDR

Materials discovery can be accelerated by learning from existing knowledge and data. This study demonstrates that machine learning models trained on quantum‑mechanical calculations and chemical similarity can efficiently and accurately predict a wide range of material properties. The authors formulate a general approach using one‑dimensional chain systems to derive decision rules that map easily accessible attributes to material properties. Fingerprints derived from chemo‑structural or electronic charge‑density information enable ultra‑fast, accurate predictions, and this capability substantially speeds the exploration of large chemical spaces for application‑specific materials.

Abstract

The materials discovery process can be significantly expedited and simplified if we can learn effectively from available knowledge and data. In the present contribution, we show that efficient and accurate prediction of a diverse set of properties of material systems is possible by employing machine (or statistical) learning methods trained on quantum mechanical computations in combination with the notions of chemical similarity. Using a family of one-dimensional chain systems, we present a general formalism that allows us to discover decision rules that establish a mapping between easily accessible attributes of a system and its properties. It is shown that fingerprints based on either chemo-structural (compositional and configurational information) or the electronic charge density distribution can be used to make ultra-fast, yet accurate, property predictions. Harnessing such learning paradigms extends recent efforts to systematically explore and mine vast chemical spaces, and can significantly accelerate the discovery of new application-specific materials.