Paper Image

Learning materials relationships from data fusion

Paper Authors:

A. Gilad Kusne,

Austin McDannald,

Brian DeCost


Key Details

SAGE algorithm combines multimodal data through Bayesian coregionalization

Learns synthesis-structure-property relationships from multiple data sources

Quantifies uncertainty and exploits shared trends and discontinuities

Can help overcome challenges of complex search spaces and data fusion

Demonstrated for materials science, generalizable to other fields

AI generated summary

Learning materials relationships from data fusion

This paper presents an algorithm called SAGE that combines data from multiple sources to learn the relationships between material synthesis methods, structure, and properties. SAGE uses a Bayesian approach and multimodal coregionalization to merge data and quantify uncertainty. This allows exploiting shared trends and discontinuities across data sources. SAGE can help overcome challenges of high-dimensional search spaces and data fusion to advance materials discovery.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up