.png){kind=icon}
Return
Constructing generalizable microstructure-property maps across diverse microstructure classes
.png){kind=icon}
OA
AI
DOI：10.1557/s43579-025-00873-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Establishing microstructure-property correlations that generalize across diverse microstructural classes remains a critical challenge in data-driven materials design. In this work, we evaluate the potential to extrapolate predictive models trained on one microstructure type (e.g., spinodal) to others (e.g., dendritic), using three distinct featurization strategies: two-point correlation functions, graph-based descriptors, and deep neural network embeddings. Our findings reveal that the Wasserstein distance is an excellent metric that correlates well with generalizability, serving as a model-agnostic yet data-aware signature of generalizability. Furthermore, we demonstrate that featurizations that conserve key microstructural features generalize better.
Keywords:
Microstructure
Machine Learning
Journal
IF:
2.3
Papers:
244
Citations:
2.8K
Organization
Cited Papers
ShareSaveShareSaveMachine learning of microstructure–property relationships in materials leveraging microstructure representation from foundational vision transformers.png)
Acta Materialia
Acta MaterialiaIF9.3
PREAI
ShareSaveData Infrastructure Elements in Support of Accelerated Materials Innovation: ELA, PyMKS, and MATIN
PREAI
ShareSaveShareSaveGraSPI: Extensible software for the graph-based quantification of morphology in organic electronicsSoftwareX
OAAI
SoftwareXIF0
OAAIShareSave
View more