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.
Machine learning (ML) has emerged as a powerful tool for studying the properties of condensed matter. To date, most research has focused on the bulk properties of solids, however, defects are ubiquitous in crystalline systems. Many modern functional technologies are constrained or enabled by their presence. For example, in photovoltaics, point-defects can introduce electronic states that fall within the band gap and result in voltage losses through trapping and non-radiative recombination. In contrast, point-defect qubits are an emerging platform for quantum computing and sensing that are uniquely enabled by long lived and addressable spin states on paramagnetic point defects. Defects pose a particular challenge for modern machine learning methods since they are often charged, leading to long range forces that are not well captured by existing approaches. This collection focuses on the development and application of novel machine learning approaches to study the geometric, thermodynamic, kinetic, and electronic properties of defects in the solid state.