Machine learning prediction of magnetic properties of Fe-based metallic glasses considering glass forming ability

Li, Xin; Shan, Guangcun; Shek, C.H.

doi:10.1016/j.jmst.2021.05.076

Cited by 41 publications

(9 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, many research articles in the materials science field have clearly pointed out that the XGBoost algorithm has superior performance over other algorithms. [131][132][133]…”

Section: Gradient Boostingmentioning

confidence: 99%

Methods, progresses, and opportunities of materials informatics

Li,

Zheng

2023

InfoMat

View full text Add to dashboard Cite

As an implementation tool of data intensive scientific research methods, machine learning (ML) can effectively shorten the research and development (R&D) cycle of new materials by half or even more. ML shows great potential in the combination with other scientific research technologies, especially in the processing and classification of large amounts of material data from theoretical calculation and experimental characterization. It is very important to systematically understand the research ideas of material informatics to accelerate the exploration of new materials. Here, we provide a comprehensive introduction to the most commonly used ML modeling methods in material informatics with classic cases. Then, we review the latest progresses of prediction models, which focus on new processing–structure–properties–performance (PSPP) relationships in some popular material systems, such as perovskites, catalysts, alloys, two‐dimensional materials, and polymers. In addition, we summarize the recent pioneering researches in innovation of material research technology, such as inverse design, ML interatomic potentials, and microtopography characterization assistance, as new research directions of material informatics. Finally, we comprehensively provide the most significant challenges and outlooks related to the future innovation and development in the field of material informatics. This review provides a critical and concise appraisal for the applications of material informatics, and a systematic and coherent guidance for material scientists to choose modeling methods based on required materials and technologies.

show abstract

“…Recently, many research articles in the materials science field have clearly pointed out that the XGBoost algorithm has superior performance over other algorithms. [131][132][133]…”

Section: Gradient Boostingmentioning

confidence: 99%

Methods, progresses, and opportunities of materials informatics

Li,

Zheng

2023

InfoMat

View full text Add to dashboard Cite

show abstract

“…42 The optimized MLP architecture had two hidden layers and twelve nodes in each hidden layer. In addition, 10-fold cross-validation 43 was conducted for each combination of hyperparameters. As mentioned above, RMSE was used for the evaluation of the predictive performance of the trained ML models.…”

Section: Machine Learning and Multi-objective Optimizationmentioning

confidence: 99%

Accelerated design for magnetic high entropy alloys using data-driven multi-objective optimization

Shan

Zhang

et al. 2022

J. Mater. Chem. C

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this study, three ML methods, namely, random forest (RF), extreme gradient boosting (XGBoost), and convolutional neural network (CNN), were considered to develop the highest accuracy model for predicting the PR of graphene structures. These ML methods have been tried in several previous studies regarding the optimization of material synthesis or the investigation of material properties ,, and others . The XGBoost is a decision-tree-based ensemble ML algorithm that was developed by Chen and Guestrin in 2016 .…”

Section: Ml-based Optimization Of Hydrogenated Graphene Structurementioning

confidence: 99%

Hydrogenated Graphene with Tunable Poisson’s Ratio Using Machine Learning: Implication for Wearable Devices and Strain Sensors

Nguyen

Nguyễn

et al. 2022

ACS Appl. Nano Mater.

View full text Add to dashboard Cite

The Poisson's ratio of two-dimensional materials such as graphene can be tailored by surface hydrogenation. The density and distribution of hydrogenation may significantly affect the Poisson's ratio of the graphene structure. Therefore, optimization of the distribution of hydrogenation is useful to achieve the structure with a targeted Poisson's ratio. For this purpose, we developed an inverse design algorithm based on machine learning using the XGBoost method to reveal the relationship between the Poisson's ratio and distribution of hydrogenation. Based on this relationship, we can optimize the hydrogenated graphene structure to have a low Poisson's ratio. Instead of performing molecular dynamic simulations for all possible structures, we could find the optimal structures using the search algorithm and save significant computational resources. This algorithm could successfully discover structures with low Poisson's ratios around −0.5 after only 1600 simulations in a large design space of approximately 5.2 × 10 6 possible configurations. Moreover, the optimal structures were found to exhibit excellent flexibility under compression of around −65% without failure and can be used in many applications such as flexible strain sensors. Our results demonstrate the applicability of machine learning to the efficient development of new metamaterials with desired properties.

show abstract

Machine learning prediction of magnetic properties of Fe-based metallic glasses considering glass forming ability

Cited by 41 publications

References 55 publications

Methods, progresses, and opportunities of materials informatics

Methods, progresses, and opportunities of materials informatics

Accelerated design for magnetic high entropy alloys using data-driven multi-objective optimization

Hydrogenated Graphene with Tunable Poisson’s Ratio Using Machine Learning: Implication for Wearable Devices and Strain Sensors

Contact Info

Product

Resources

About