The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index.
A critical examination of compound stability predictions from machine-learned formation energies
Open accessAbstract
Machine learning has emerged as a novel tool for the efficient prediction of material properties, and claims have been made that machine-learned models for the formation energy of compounds can approach the accuracy of Density Functional Theory (DFT). The models tested in this work include five recently published compositional models, a baseline model using stoichiometry alone, and a structural model. By testing seven machine learning models for formation energy on stability predictions using the Materials Project database of DFT calculations for 85,014 unique chemical compositions, we show that while formation energies can indeed be predicted well, all compositional models perform poorly on predicting the stability of compounds, making them considerably less useful than DFT for the discovery and design of new solids. Most critically, in sparse chemical spaces where few stoichiometries have stable compounds, only the structural model is capable of efficiently detecting which materials are stable. The nonincremental improvement of structural models compared with compositional models is noteworthy and encourages the use of structural models for materials discovery, with the constraint that for any new composition, the ground-state structure is not known a priori. This work demonstrates that accurate predictions of formation energy do not imply accurate predictions of stability, emphasizing the importance of assessing model performance on stability predictions, for which we provide a set of publicly available tests.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Although this simplified GNN architecture is able to achieve state-of-the-art generalization on the Materials Project (MAE of 60 meV atom−1 (ref. 25)), it does not offer useful predictions for materials discovery, which was also observed by Bartel et al.25.”
From Scaling deep learning for materials discovery (2023) · cites this paper as [25]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Although data-driven methods that aid in further materials discovery have been pursued, thus far, machine-learning techniques have been ineffective in estimating stability (decomposition energy) with respect to the convex hull of energies from competing phases25.”
From Scaling deep learning for materials discovery (2023) · cites this paper as [25]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“The Li/Mn transition-metal oxide family is discussed in ref. 25 to analyse the capabilities of machine-learning models for use in discovery. In the main text, we compare against the findings in the cited work suggesting limited discovery within this family through previous machine-learning methods.”
From Scaling deep learning for materials discovery (2023) · cites this paper as [25]
The cited source says what the claim attributes to it.
In the citing paper
“Although this simplified GNN architecture is able to achieve state-of-the-art generalization on the Materials Project (MAE of 60 meV atom−1 (ref. 25)), it does not offer useful predictions for materials discovery, which was also observed by Bartel et al.25.”
From Scaling deep learning for materials discovery (2023) · cites this paper as [25]
The cited source says what the claim attributes to it.
In the citing paper
“Finally, Li/Mn transition-metal oxides are a promising family to replace LiCoO2 in rechargeable batteries25 and GNoME has discovered an extra 15 candidates stable relative to the Materials Project compared with the original nine.”
From Scaling deep learning for materials discovery (2023) · cites this paper as [25]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
XGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation
Jain, A., Ong, S. P., Hautier, G., Chen, W., Richards, W. D., Dacek, S., Cholia, S., Gunter, D., Skinner, D., Ceder, G., et al. (2013) · APL Materials · doi:10.1063/1.4812323 · Cited by 13,094
Self-citationShow, Attend and Tell: Neural Image Caption Generation with Visual Attention
Xu, K., Ba, J., Kiros, R., Cho, K., Courville, A., Salakhutdinov, R., Zemel, R., & Bengio, Y. (2015) · arXiv · doi:10.48550/arxiv.1502.03044 · Cited by 8,248
Machine learning for molecular and materials science
Butler, K. T., Davies, D. W., Cartwright, H., Isayev, O., & Walsh, A. (2018) · Nature · doi:10.1038/s41586-018-0337-2 · Cited by 4,968
Xie, T., & Grossman, J. C. (2018) · Physical Review Letters · doi:10.1103/physrevlett.120.145301 · Cited by 2,828
Oxidation energies of transition metal oxides within the GGA + U framework
Wang, L., Maxisch, T., & Ceder, G. (2006) · Physical Review B · doi:10.1103/physrevb.73.195107 · Cited by 2,785
Self-citationRecent advances and applications of machine learning in solid-state materials science
Schmidt, J., Marques, M. R. G., Botti, S., & Marques, M. A. L. (2019) · npj Computational Materials · doi:10.1038/s41524-019-0221-0 · Cited by 2,607
On representing chemical environments
Bartók, A. P., Kondor, R., & Csányi, G. (2013) · Physical Review B · doi:10.1103/physrevb.87.184115 · Cited by 2,572
The Open Quantum Materials Database (OQMD): assessing the accuracy of DFT formation energies
Kirklin, S., Saal, J. E., Meredig, B., Thompson, A., Doak, J. W., Aykol, M., Rühl, S., & Wolverton, C. (2015) · npj Computational Materials · doi:10.1038/npjcompumats.2015.10 · Cited by 2,378
Li2MnO3-stabilized LiMO2 (M = Mn, Ni, Co) electrodes for lithium-ion batteries
Thackeray, M. M., Kang, S.-H., Johnson, C. S., Vaughey, J. T., Benedek, R., & Hackney, S. A. (2007) · Journal of Materials Chemistry · doi:10.1039/b702425h · Cited by 2,039
Research works printed but not linked: 6
- Choudhary, K., Kalish, I., Beams, R. & Tavazza, F. High-throughput identification and characterization of two-dimensional materials using density functional theory. Sci. Rep. 7, 5179 (2017).
- Davies, D. W. et al. Computational screening of all stoichiometric. Inorg. Mater. Chem. 1, 617–627 (2016).
- Dunn, A., Wang, Q., Ganose, A., Dopp, D. & Jain, A. Benchmarking materials property prediction methods: the Matbench test set and automatminer reference algorithm. Preprint at https://arxiv.org/abs/2005.00707. · arXiv
- Goodall, R., & Lee, A. Predicting materials properties without crystal structure: deep representation learning from stoichiometry. https://arxiv.org/abs/1910.00617. · arXiv
- Zhang, Y. et al. Efficient first-principles prediction of solid stability: towards chemical accuracy. npj Comput. Mater. 4, 9 (2018).
- Miedema, A. R. Simple model for alloys. Philips Tech. Rev. 33(no. 6), 149–160 (1973).
Cited by
Scaling deep learning for materials discovery
Merchant, A., Batzner, S., Schoenholz, S. S., Aykol, M., Cheon, G., & Cubuk, E. D. (2023) · Nature · doi:10.1038/s41586-023-06735-9 · Cited by 1,527
“Although data-driven methods that aid in further materials discovery have been pursued, thus far, machine-learning techniques have been ineffective in estimating stability (decomposition energy) with respect to the convex hull of energies from competing phases 25 .”
BackgroundMethodInfluentialRecent advances and applications of deep learning methods in materials science
Choudhary, K., DeCost, B., Chen, C., Jain, A., Tavazza, F., Cohn, R., Park, C. W., Choudhary, A., Agrawal, A., Billinge, S. J. L., et al. (2022) · npj Computational Materials · doi:10.1038/s41524-022-00734-6 · Cited by 1,172
“It is thus not a surprise that chemical formula-based methods may not be adequate in some cases 159 .”
BackgroundSelf-citationGraph neural networks for materials science and chemistry
Reiser, P., Neubert, M., Eberhard, A., Torresi, L., Zhou, C., Shao, C., Metni, H., van Hoesel, C., Schopmans, H., Sommer, T., et al. (2022) · Communications Materials · doi:10.1038/s43246-022-00315-6 · Cited by 860
Compositionally restricted attention-based network for materials property predictions
Wang, A. Y.-T., Kauwe, S. K., Murdock, R. J., & Sparks, T. D. (2021) · npj Computational Materials · doi:10.1038/s41524-021-00545-1 · Cited by 260
“CritExam__Ed Bartel et al.(55) Decomposition enthalpy 85014 (70/15/15)”
BackgroundReview of computational approaches to predict the thermodynamic stability of inorganic solids
Bartel, C. J. (2022) · Journal of Materials Science · doi:10.1007/s10853-022-06915-4 · Cited by 196
Self-citation
Metadata
- Authors
- Bartel, C. J., Trewartha, A., Wang, Q., Dunn, A., Jain, A., & Ceder, G.
- Affiliations
- University of California, Berkeley (US)
- Lawrence Berkeley National Laboratory (US)
- Corresponding author
- Bartel, Christopher J.
- Ceder, Gerbrand
- Venue
- npj Computational Materials
- Published
- 2020-07-10
- Volume
- 6
- Issue
- 1
- ISSN
- 2057-3960 (ISSN-L)
- OpenAlex
- W3041359432
- Cited by
- 277
- References
- 43
- Citations by year
- 2026: 65 · 2025: 54 · 2024: 47 · 2023: 40 · 2022: 41 · 2021: 26 · 2020: 4
- FWCI
- 10.79
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- npj Computational Materials (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- eScholarship (California Digital Library) (submittedVersion)
- OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
- arXiv (Cornell University)
- Funders
- U.S. Department of Energy (DE-AC02-05-CH11231)
- Topics
- Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
- Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Advanced Chemical Physics Studies
- Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
- Keywords
- formation energy, machine learning, compositional models, structural model, Materials Project, stoichiometry, materials discovery, model evaluation, ground state structure
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A critical examination of compound stability predictions from machine-learned formation energies. https://app.cochat.ai/papers/10.1038/s41524-020-00362-y
- The paper
- Bartel, C. J., Trewartha, A., Wang, Q., Dunn, A., Jain, A., & Ceder, G. (2020). A critical examination of compound stability predictions from machine-learned formation energies. npj Computational Materials. https://doi.org/10.1038/s41524-020-00362-y
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.