MLAMD Center AI/ML + exascale computing

Publications

Papers, preprints and software releases

Publications and manuscript outputs that connect workflow development to magnetic and functional materials applications.

Outputs

Papers, preprints and software releases

In progress

Discovery of novel magnetic Y-Mn-B compounds via an advanced machine-learning-guided framework

W. Xia et al. Manuscript in progress.

Published paper

Accelerated discovery and design of Fe-Co-Zr magnets with tunable magnetic anisotropy through machine learning and parallel computing

W. Xia, M. Moraru, Y. W. Li, T. Liao, J. R. Chelikowsky, and C. Z. Wang. Physical Review Materials 10(2), 024403 (2026).

Read the paper

Preprint

exaPD: A highly parallelizable workflow for multi-element phase diagram (PD) construction

F. Zhang, Z. Ye, M. Moraru, Y. W. Li, W. Xia, Y. Yao, R. Richard, and C. Z. Wang. arXiv:2510.01400 (2025).

Read the preprint

Preprint

exa-AMD: An Exascale-Ready Framework for Accelerating the Discovery and Design of Functional Materials

W. Xia, M. Moraru, Y. W. Li, and C. Z. Wang. arXiv:2510.01170 (2025).

Read the preprint

Software paper

exa-AMD: A Scalable Workflow for Accelerating AI-Assisted Materials Discovery and Design

M. Moraru, W. Xia, Z. Ye, F. Zhang, Y. Yao, Y. W. Li, and C. Z. Wang. Journal of Open Source Software 10(115), 8879.

Read the JOSS paper
Five-stage exa-AMD workflow from structure generation to post-processing
exa-AMD workflow: structure generation, ML screening, selection, first-principles calculations, and post-processing.
Neural network potential schematic for molecular dynamics simulations
Neural network potentials enable exa-PD to extend atomistic simulation length and time scales.