Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Editor's choice: Machine learning for material discovery, design and characterisation
Machine learning has been rapidly developing in the past decade and has become an important tool to the scientific community. Publicly available datasets can be exploited to automate the discovery of new materials, bringing unlikely solutions to material problems to our attention in a way that human researchers might be unable, or unlikely, to spot. Machine learning techniques can be applied not only to the discovery of materials, but also to design devices or characterise materials from images or limited datasets. Here, we present a number of publications in Scientific Reports that highlight the achievements of researchers in this area.