Zitieren Sie bitte immer diesen URN: urn:nbn:de:kobv:b43-637139

A framework to evaluate machine learning crystal stability predictions

  • The rapid adoption of machine learning in various scientific domains calls for the development of best practices and community agreed-upon benchmarking tasks and metrics. We present Matbench Discovery as an example evaluation framework for machine learning energy models, here applied as pre-filters to first-principles computed data in a high-throughput search for stable inorganic crystals. We address the disconnect between (1) thermodynamic stability and formation energy and (2) retrospective and prospective benchmarking for materials discovery. Alongside this paper, we publish a Python package to aid with future model submissions and a growing online leaderboard with adaptive user-defined weighting of various performance metrics allowing researchers to prioritize the metrics they value most. To answer the question of which machine learning methodology performs best at materials discovery, our initial release includes random forests, graph neural networks, one-shot predictors,The rapid adoption of machine learning in various scientific domains calls for the development of best practices and community agreed-upon benchmarking tasks and metrics. We present Matbench Discovery as an example evaluation framework for machine learning energy models, here applied as pre-filters to first-principles computed data in a high-throughput search for stable inorganic crystals. We address the disconnect between (1) thermodynamic stability and formation energy and (2) retrospective and prospective benchmarking for materials discovery. Alongside this paper, we publish a Python package to aid with future model submissions and a growing online leaderboard with adaptive user-defined weighting of various performance metrics allowing researchers to prioritize the metrics they value most. To answer the question of which machine learning methodology performs best at materials discovery, our initial release includes random forests, graph neural networks, one-shot predictors, iterative Bayesian optimizers and universal interatomic potentials. We highlight a misalignment between commonly used regression metrics and more task-relevant classification metrics for materials discovery. Accurate regressors are susceptible to unexpectedly high false-positive rates if those accurate predictions lie close to the decision boundary at 0 eV per atom above the convex hull. The benchmark results demonstrate that universal interatomic potentials have advanced sufficiently to effectively and cheaply pre-screen thermodynamic stable hypothetical materials in future expansions of high-throughput materials databases.zeige mehrzeige weniger

Volltext Dateien herunterladen

Metadaten exportieren

Weitere Dienste

Suche bei Google Scholar Anzahl der Zugriffe auf dieses Dokument
Metadaten
Autor*innen:Janosh RiebesellORCiD, Rhys E. A. GoodallORCiD, Philipp BennerORCiD, Yuan Chiang, Bowen DengORCiD, Gerbrand CederORCiD, Mark Asta, Alpha A. Lee, Anubhav JainORCiD, Kristin A. PerssonORCiD
Dokumenttyp:Zeitschriftenartikel
Veröffentlichungsform:Verlagsliteratur
Sprache:Englisch
Titel des übergeordneten Werkes (Englisch):Nature Machine Intelligence
Jahr der Erstveröffentlichung:2025
Organisationseinheit der BAM:VP Vizepräsident
VP Vizepräsident / VP.1 eScience
Veröffentlichende Institution:Bundesanstalt für Materialforschung und -prüfung (BAM)
Verlag:Springer Science and Business Media LLC
Jahrgang/Band:7
Ausgabe/Heft:6
Erste Seite:836
Letzte Seite:847
DDC-Klassifikation:Technik, Medizin, angewandte Wissenschaften / Ingenieurwissenschaften / Ingenieurwissenschaften und zugeordnete Tätigkeiten
Freie Schlagwörter:Interatomic Potentials; Machine Learning
Themenfelder/Aktivitätsfelder der BAM:Material
Material / Materialdesign
DOI:10.1038/s42256-025-01055-1
URN:urn:nbn:de:kobv:b43-637139
ISSN:2522-5839
Verfügbarkeit des Dokuments:Datei für die Öffentlichkeit verfügbar ("Open Access")
Lizenz (Deutsch):License LogoCreative Commons - CC BY - Namensnennung 4.0 International
Datum der Freischaltung:17.07.2025
Referierte Publikation:Ja
Datum der Eintragung als referierte Publikation:17.07.2025
Schriftenreihen ohne Nummerierung:Wissenschaftliche Artikel der BAM
Einverstanden
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.