Large-scale prediction of new stable crystalline materials (GNoME)
GNoME predicted about 2.2 million new inorganic crystal structures, roughly 380,000 of them flagged as thermodynamically stable, presented as an order-of-magnitude expansion of known stable materials.
- Model
- GNoME (graph neural network)
- Field
- Materials science
- Date
- 2023-11-29
What was found
Graph Networks for Materials Exploration (GNoME) is a graph neural network trained in an active-learning loop against density-functional-theory calculations to predict formation energies and filter candidates by stability. DeepMind released the predicted structures as a public dataset. Published in Nature.
Novelty check
Stable-material prediction builds on the Materials Project and OQMD convex-hull databases. The dispute below is precisely about how much of GNoME's set is genuinely new relative to those existing catalogues.
Caveats and known objections
Substantive objections were raised and remain unresolved. Cheetham and Seshadri, reviewing the release, found 'scant evidence for compounds that fulfil the trifecta of novelty, credibility and utility,' noting that many entries are minor substitutional variants, radioactive, or otherwise unlikely to be useful. Stability here means a DFT convex-hull prediction, not experimental synthesis; the headline counts are predicted, not made.
Independent checks
Cheetham & Seshadri (critical review): disputed the novelty and utility of most predicted compounds · link ↗
Disagree with these grades?
Bring a citation: a grade moves on evidence, not on argument.
Or on GitHub: submit a check challenge the grade send a correction or send a pull request
Flag this for triage
Signals order the review queue and nothing else. They are never published, and they never move a grade: that takes a citation.
Entry history (1 event)
- AddedEntered the registry graded Disputed and Search scaffold.
Entries are never deleted. A grade that does not hold up is downgraded on the record, with the reason beside it.
Community discussion
Graded disputed for verification and search scaffold for autonomy. What these mean.
Cite this entry
whataifound.org. (2023). Large-scale prediction of new stable crystalline materials (GNoME). whataifound.org: A Registry of AI Scientific and Mathematical Discoveries. https://whataifound.org/finding/2023-11-29-gnomeBibTeX
@misc{whataifound-googledeepmind-2023-gnome,
title = {Large-scale prediction of new stable crystalline materials (GNoME)},
author = {{whataifound.org}},
year = {2023},
howpublished = {whataifound.org: A Registry of AI Scientific and Mathematical Discoveries},
note = {Result by Google DeepMind. Verification: Disputed. Autonomy: Search scaffold.},
url = {https://whataifound.org/finding/2023-11-29-gnome}
}