Browse Teach Curate

Machine Learning Prediction of Exciton Binding Energies in 2D Materials

A data-driven surrogate for expensive GW-Bethe-Salpeter calculations: supervised regressors trained on a computational 2D-materials database predict quasiparticle band gaps and exciton binding energies from cheap structural and electronic descriptors, and Bayesian optimization screens for materials with the strongest excitonic effects.

Rosie Text

Machine Learning-Driven Insights into Excitonic Effects in 2D Materials Ahsan Javed1,2, Sajid Ali∗,3

Exciton binding energy (EBE) is the energy needed to split a bound electron-hole pair in a material into free charge carriers, and is normally obtained from expensive first-principles many-body calcu…