Data-driven assessment of thermodynamic stability and search for competing phases: Application to the 2D Material Defect dataset

Roman A. Eremin, Aliaksei V. Krautsou, Innokentiy S. Humonen, Artem D. Dembitskiy, Semen Budennyy

Keywords: machine learning, density functional theory, graph neural networks, chemically modified crystals

TL;DR: for the 2DMD dataset, an approach to constructing a set of possible competing phases and assessing their impact on thermodynamic properties is described

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 75