Rare-earth-free Fe-Co-S magnet candidates discovered through ML-guided high-throughput search

1/ AI use method: machine-learning-guided high-throughput first-principles search
2/ Applied task: searching for rare-earth-free magnetic material candidates in the Fe-Co-S system
3/ Reported result: presented discovery of rare-earth-free magnet candidates through ML-guided search
4/ A PRX Materials paper accepted on 2026-10-02 reported a study that searched for rare-earth-free magnet candidates in the Fe-Co-S system using machine-learning-guided high-throughput first-principles search.
5/ The core of the study is to use machine learning to prioritize the search and combine it with high-throughput first-principles calculations to find candidate materials.
6/ For materials researchers, this approach shows a typical workflow in which ML is used as a screening guide to reduce computational cost when the composition space is broad and rare elements are to be avoided.
7/ In particular, for problems such as magnetic composition search, where there are many candidates and property calculations are expensive, the combination of high-throughput search and first-principles calculations is practical.
8/ However, abstract-level information alone does not allow sufficient judgment of the model input features, the training data range, or whether the final candidates were experimentally validated.
9/ Therefore, it cannot be concluded that the same performance will be obtained in other systems, and in practical use the bias of the data and the risk of extrapolation should be checked first. [1]