Discovery of a material with low thermal expansion, high Curie temperature, and phase stability by finding candidate multicomponent Invar alloys through active learning

AI editorPublished 2026.09.11Posted 2026.09.11 18:01 KST

It was reported that by using active learning to explore the compositions of multicomponent Invar alloys and screen candidates, a material with low thermal expansion, high Curie temperature, and excellent phase stability was discovered. This can be interpreted as a case in which active learning was used to reduce the candidate space in materials discovery. [1]

1/ AI 활용 방법: Active learning
2/ 적용 과제: Invar multicomponent alloy composition exploration and candidate screening
3/ 보고된 결과: It was reported that a multicomponent Invar material with low thermal expansion, high Curie temperature, and excellent phase stability was discovered.
4/ This study reported that active learning was used to explore the composition space of Invar multicomponent alloys and screen promising candidates, leading to the discovery of a material with low thermal expansion, high Curie temperature, and excellent phase stability.
5/ From a materials perspective, the significance is that rather than testing a broad composition space blindly, the model selected the next experimental candidates, improving exploration efficiency.
6/ In problems with large combinations such as multicomponent alloys, this approach can be useful for saving experimental and computational resources while finding compositions with target properties.
7/ However, this result is based on a specific Invar family composition space and experimental/model settings, so it cannot be directly generalized to other alloy systems or other performance targets.
8/ In addition, the effectiveness of active learning is highly dependent on the range and quality of the initial data, the definition of the candidate space, and the evaluation criteria. [1]

How to apply this

Use active learning to prioritize candidate compositions in multicomponent alloys and reduce the number of experiments in exploration.

Train a composition-property prediction model using initial measurement or computational data.
Select the next composition to experiment with using prediction uncertainty or an acquisition function.
Synthesize or compute the selected composition and evaluate thermal expansion, Curie temperature, and phase stability.
Add the new results to the dataset, update the model, and select the next candidate again.

Data needed: Initial alloy composition-property data, the composition candidate space, and sample or computational results that can evaluate thermal expansion, Curie temperature, and phase stability for each candidate are needed.
Checking the result: Check whether the measured values of newly recommended compositions actually improve over the previous data, and see whether the same trend is reproduced in independent validation samples or repeated experiments.
What can go wrong: If the initial data are narrow or biased, the model may miss promising regions or repeatedly recommend overly similar compositions.

이 가이드는 원문의 방법을 바탕으로 AI가 제안한 적용 방향입니다. 원문에 적힌 내용이 아니며 같은 결과가 재현된다는 보장도 아닙니다.

웹 자료를 바탕으로 AI가 자동 요약한 글입니다. 연구의 세부 조건과 결과는 연결된 원문을 확인해주세요.

Discovery of a material with low thermal expansion, high Curie temperature, and phase stability by finding candidate multicomponent Invar alloys through active learning

How to apply this

Related concepts and practice

Sources