Indirect prediction and screening of frustrated magnetic materials using machine learning
Abstract
Magnetic frustration is commonly quantified by the frustration index, calculated from the Curie-Weiss and Neel temperatures. Because both quantities are not simultaneously available for most materials, direct prediction of the magnetic frustration index remains challenging. Here, we propose a machine learning workflow for indirect prediction of the frustration index and consequent screening of potential frustrated magnetic compounds through independent prediction of Curie-Weiss and Neel temperatures. In our work, composition-based models implemented via the CatBoost algorithm, trained on experimental data were optimized using chemically informed dataset subdivision. The highest predictive accuracy was achieved for the rare earth-oxide model in predicting Neel and for the rare earth model in predicting Curie-Weiss. Ensemble uncertainty was incorporated through a conservative frustration index to improve the reliability of candidate selection. Screening of approximately 67000 Materials Project compounds identified 20 high-confidence frustrated magnetic candidates, including several experimentally known frustrated magnets and previously unexplored oxide systems potential for further investigation.
Related
- Vortex order in magnetic frustrated GeNi2O4 and GeCo2O4 spinels
- Frustrated Gd3+ Double Perovskites as High-Performance Magnetocaloric Materials for Sub-100 mK Adiabatic Demagnetization Refrigeration
- Triplon dynamics from partial Kondo screening in frustrated heavy-fermion systems
- Mechanism of Incommensurate Magnetic Order in BaCo2(AsO4)2: Interplay of Frustrated Further-Neighbor Exchanges and Bond-Directional Anisotropy
- The bixbyite framework as a platform for frustrated noncollinear magnetism: resolving the magnetic ground state of β-Fe2O3
BibTeX
@misc{burganova2026indirect,
title = {{Indirect prediction and screening of frustrated magnetic materials using machine learning}},
author = {Regina Burganova and Irina Piyanzina and Ruslan Akhmerov and Alexander Sigov and Maxim Fedorov},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.15683},
eprint = {2610.15683},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.15683},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Indirect prediction and screening of frustrated magnetic materials using machine learning
AU - Regina Burganova
AU - Irina Piyanzina
AU - Ruslan Akhmerov
AU - Alexander Sigov
AU - Maxim Fedorov
PY - 2026
DA - 2026/10/11
PB - arXiv
DO - 10.48550/arxiv.2610.15683
UR - https://arxiv.org/abs/2610.15683
AB - Magnetic frustration is commonly quantified by the frustration index, calculated from the Curie-Weiss and Neel temperatures. Because both quantities are not simultaneously available for most materials, direct prediction of the magnetic frustration index remains challenging. Here, we propose a machine learning workflow for indirect prediction of the frustration index and consequent screening of potential frustrated magnetic compounds through independent prediction of Curie-Weiss and Neel temperatures. In our work, composition-based models implemented via the CatBoost algorithm, trained on experimental data were optimized using chemically informed dataset subdivision. The highest predictive accuracy was achieved for the rare earth-oxide model in predicting Neel and for the rare earth model in predicting Curie-Weiss. Ensemble uncertainty was incorporated through a conservative frustration index to improve the reliability of candidate selection. Screening of approximately 67000 Materials Project compounds identified 20 high-confidence frustrated magnetic candidates, including several experimentally known frustrated magnets and previously unexplored oxide systems potential for further investigation.
ER -