Feature optimization for atomistic machine learning yields a data-driven construction of the periodic table of the elements
Open accessAbstract
By representing elements as points in a low-dimensional chemical space it is possible to improve the performance of a machine-learning model for a chemically-diverse dataset. The resulting coordinates are reminiscent of the main groups of the periodic table.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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 accuracy of the regression model provides an objective metric to optimize the hyperparameters that define the representation, and even to determine the most effective non-linear function of features that reproduces the target properties.
“The accuracy of the regression model also provides an objective metric to optimize the hyperparameters that define the representation 107 , and even to determine the most effective non-linear function of features that reproduces the target properties 108 .”
From Unsupervised machine learning in atomistic simulations, between predictions and understanding (2019) · cites this paper as [107]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 46 · Cited by 141
Research works printed but not linked: 14
- [11]D. Dragoni, T. D. Daff, G. Csanyi, and N. Marzari, Physical Review Materials 2 , 013808 (2017).
- [14]S. De, F. Musil, T. Ingram, C. Baldauf, and M. Ceriotti, Journal of Cheminformatics 9 , 6 (2017).
- [19]J. Behler, Journal of Chemical Physics 134 , 074106 (2011).
- [22]L. Zhang, J. Han, H. Wang, R. Car, and E. Weinan, Physical Review Letters 120 , 143001 (2018).
- [24]C. Qu, Q. Yu, B. L. Van Hoozen, J. M. Bowman, and R. A. Vargas-Hern´ andez, Journal of Chemical Theory and Computation 14 , 3381 (2018).
- [25]C. E. Rasmussen, Gaussian processes for machine learning , MIT Press, 2006.
- [26]C. M. Bishop, Pattern Recognition and Machine Learning , Springer, 2016.
- [28]M. J. Willatt, F. Musil, and M. Ceriotti, Arxiv (2018).
- [30]F. M. Paruzzo et al., arXiv preprint arXiv:1805.11541 (2018). · arXiv
- [31]F. A. Faber, A. S. Christensen, B. Huang, and O. A. Von Lilienfeld, Journal of Chemical Physics 148 , 241717 (2018).
- [37]I. Goodfellow, Y. Bengio, A. Courville, and Y. Bengio, Deep learning , volume 1, MIT press Cambridge, 2016.
- [45]Q. Zhou et al., Proceedings of the National Academy of Sciences 115 , E6411 (2018).
- [47]J. Barker, J. Bulin, J. Hamaekers, and S. Mathias, LCGAP: Localized coulomb descriptors for the gaussian approximation potential, in Scientific Computing and Algorithms in Industrial Simulations: Projects and Products of Fraunhofer SCAI , pages 25-42, 2017.
- [49]F. A. Faber et al., Journal of Chemical Theory and Computation 13 , 5255 (2017).
- Authors
- Willatt, M. J., Musil, F., & Ceriotti, M.
- Affiliations
- École Polytechnique Fédérale de Lausanne (CH)
- Corresponding author
- Willatt, Michael J.
- ORCID
- Ceriotti, Michele
- Venue
- Physical Chemistry Chemical Physics
- Publisher
- Royal Society of Chemistry (RSC)
- Published
- 2018
- Volume
- 20
- Issue
- 47
- Pages
- 29661-29668
- ISSN
- 1463-9076 (ISSN-L)
- 1463-9084
- OpenAlex
- W2900483837
- Cited by
- 141
- References
- 46
- Citations by year
- 2026: 11 · 2025: 15 · 2024: 18 · 2023: 18 · 2022: 20 · 2021: 21 · 2020: 20 · 2019: 17 · 2018: 1
- FWCI
- 5.49
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Physical Chemistry Chemical Physics (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Infoscience (Ecole Polytechnique Fédérale de Lausanne) (submittedVersion, cc-by-nc-nd)
- arXiv (Cornell University)
- Funders
- H2020 European Research Council (677013-HBMAP)
- Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (NCCR MARVEL)
- 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
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
- Keywords
- feature optimization, periodic table of the elements, atomistic machine learning, interatomic potentials, element correlation, multiscale interactions, molecular stability, coarse-grained potentials, structure-property relationships
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
