Can machine learning identify the next high-temperature superconductor? Examining extrapolation performance for materials discovery
Bryce Meredig, Erin Antono, Carena Church, Maxwell Hutchinson, Julia Ling, Sean P. Paradiso, Ben J. Blaiszik, Ian T Foster, Brenna M. Gibbons, Jason R. Hattrick-Simpers, Apurva Mehta, Logan T. Ward
Abstract
Traditional machine learning (ML) metrics overestimate model performance for materials discovery.
Abstract of "Can machine learning identify the next high-temperature superconductor? Examining extrapolation performance for materials discovery" by Bryce Meredig, Erin Antono, Carena Church et al., https://doi.org/10.1039/c8me00012c, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified.
Record
- Archive ID
- doi:10.1039/c8me00012c
- Published
- Jan 1, 2018
- Venue
- Molecular Systems Design & Engineering
- DOI
- 10.1039/c8me00012c
- Categories
- Machine Learning in Materials Science
- License
- cc-by
- Provenance
- openalex