2019/10/31 by Nikolas Claussen, B. Andrei Bernevig, Nicolas Regnault · 1 citation
Physics and Astronomy · #cond-mat.mtrl-sci #physics.comp-ph
paper · pdf · doi:10.1103/physrevb.101.245117
published as Phys. Rev. B 101, 245117 (2020) · 34 pages, 7 figures. The ML model is available online at https://www.topologicalquantumchemistry.com/mltqc. Version 2 includes corrections after peer review
arxiv created 2020/06/03 · arxiv updated 2020/07/01
Databases compiled using ab-initio and symmetry-based calculations now contain tens of thousands of topological insulators and topological semimetals. This makes the application of modern machine learning methods to topological materials possible. Using gradient boosted trees, we show how to construct a machine learning model which can predict the topology of a given existent material with an accuracy of 90%. Such predictions are orders of magnitude faster than actual ab-initio calculations. Through extensive testing of different models we determine which properties help detect topological materials. We identify the sources of our model's errors and we discuss approaches to overcome them.