Abstract

The computational study of materials - from predicting crystal structures to understanding lattice dynamics and electron-phonon (el-ph) coupling - has undergone a transformation in recent years. Problems that were once computationally prohibitive can now be tackled thanks to breakthrough methodologies. Machine learning interatomic potentials and Wannier function techniques, for example, have fundamentally expanded what is computationally feasible, enabling large-scale molecular dynamics, accurate phonon calculations in complex systems, and efficient el-ph coupling estimates.

This mini-colloquium will showcase recent advances across this exciting frontier. We invite contributions on topics including: crystal structure prediction using modern computational methods; ML interatomic potentials for molecular dynamics and phonon properties; anharmonic lattice dynamics and finite-temperature effects; electron-phonon coupling and superconducting properties; scalable workflows combining density functional theory with surrogate models; and advances in Wannier interpolation and related techniques.

Our goal is to bring together researchers pushing the boundaries of computational materials physics. We particularly welcome submissions showcasing applications of machine learning to these challenges, but also encourage contributions on complementary methodological advances. Early-career researchers are strongly encouraged to participate.

Organizers

Name	Affiliation
Christoph Heil	Graz University of Technology
Lilia Boeri	Sapienza University of Rome