Calculating Defects Using Machine Learned Interaction Potentials
Title: Calculating Defects Using Machine Learned Interaction Potentials
DNr: LiU-gpu-2025-14
Project Type: LiU Compute
Principal Investigator: Joel Davidsson <joel.davidsson@liu.se>
Affiliation: Linköpings universitet
Duration: 2025-12-23 – 2026-03-01
Classification: 10304
Keywords:

Abstract

I am requesting GPU resources for students who took the CDIO course (TFYA99) and need additional computational power to finalize results for a publication. The students developed a molecular dynamics (MD) code based on the MACE potential. They demonstrated that it code works in allocation LiU-compute-2025-38, but were unable to complete all the necessary computations.