A Materials Map Integrating Experimental and Computational Data via Graph-Based Machine Learning for Enhanced Materials Discovery

Y. Hashimoto, X. Jia, H. Li, T. Tomai · 2025

Read A Materials Map Integrating Experimental and Computational Data via Graph-Based Machine Learning for Enhanced Materials Discovery by Y. Hashimoto, X. Jia, H. Li, T. Tomai (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-07378.txt.

Description

Yusuke Hashimoto and co-authors build maps that join lab data and computed data on materials, in a 2025 paper. Data science promises faster discovery of new materials, but lab and computed sets are hard to merge. The team first trained a model on lab data and applied it to a computed database.

They then use a graph-based deep learning frame to draw maps linking material traits to structure. A statistical check shows the frame pulls out features that reflect how complex a structure is. Oddly, that strength did not always mean better trait forecasts. The authors link this to the model's strong learning, which can shape how points sit on the map. The maps aim to help lab researchers choose what to try next.

Catalog record

CREATORY. Hashimoto, X. Jia, H. Li, T. Tomai
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE28 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-07378

Browse the full library · About the Tower of Babel