SKM 2023 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

MM: Fachverband Metall- und Materialphysik

MM 31: Data Driven Materials Science: Big Data and Work Flows – Machine Learning

MM 31.6: Talk

Wednesday, March 29, 2023, 17:15–17:30, SCH A 251

Accelerating the Search for High-Performance, Novel Materials with Active Learning — Thomas A. R. Purcell¹, Matthias Scheffler^1,2, Luca M. Ghiringhelli^1,2, and •Christian Carbogno¹ — ¹The NOMAD Laboratory at the FHI of the Max-Planck-Gesellschaft and IRIS-Adlershof of the Humboldt-Universität zu Berlin — ²Physics Department and IRIS-Adlershof at Humboldt Universität zu Berlin, Berlin, Germany.

Active-learning frameworks have the potential to greatly accelerate the search for new materials. By balancing exploitation and exploration, these approaches can efficiently search through materials space and find the regions that are most likely to contain promising candidate materials [1]. Here we present an active learning framework, that uses an ensemble of expressions found by the sure-independence screening and sparsifying operator (SISSO) approach [2,3], and we demonstrate it for the example of discovering new thermal insulators. We statistically process the predictions of independent SISSO models to automatically select the most promising material candidates and then calculate their thermal conductivity κ_L using the ab initio Green Kubo method [4]. Using this approach we are able to find multiple new thermal insulators and gain insights into what is driving down their κ_L.

[1] A. G. Kusne, et al. Nat. Comm. 11, 5966 (2020)

[2] R. Ouyang, et al. Phys. Rev. Mater. 2, 083802 (2018)

[3] T. A. R. Purcell, et al. J. Open Source. Softw. 7, 3960 (2022)

[4] F. Knoop, et al. arXiv:2209.12720