Can machine learning identify the next high-temperature superconductor? Examining extrapolation performance for materials discovery

Traditional machine learning (ML) metrics overestimate model performance for materials discovery. We introduce (1) leave-one-cluster-out cross-validation (LOCO CV) and (2) a simple nearest-neighbor benchmark to show that model performance in discovery applications strongly depends on the problem, da...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Molecular systems design & engineering 2018-10, Vol.3 (5), p.819-825
Hauptverfasser:	Meredig, Bryce, Antono, Erin, Church, Carena, Hutchinson, Maxwell, Ling, Julia, Paradiso, Sean, Blaiszik, Ben, Foster, Ian, Gibbons, Brenna, Hattrick-Simpers, Jason, Mehta, Apurva, Ward, Logan
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial intelligence CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND SUPERFLUIDITY Data sampling Experimentation Extrapolation High temperature superconductors Iterative methods Machine learning Superconductors
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!