Accelerated Discovery of Advanced Thermoelectric Materials via Transfer Learning (Adv. Energy Mater. 23/2023)

Abstract

Thermoelectrics In article 2300049, Huijun Liu, Ying Fang, and co-workers leverage the knowledge from a pre-training on electronic bandgaps to propose a transfer learning strategy that allows ready and accurate predictions of the ZT values of any candidate materials at arbitrary temperature, which is very beneficial for the accelerated discovery of advanced thermoelectric materials.