Metraipolis › archive › dc_c19f0436dab1

Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials — arXiv:2610.09837

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.09837
License (as archived): CC0-1.0
Archived: 2026-10-08T12:37:07.865Z
SHA-256 of the archived text: 75e0c311422846f25a696db3c048dec8ce5d8ef1ca7ead95864b1800cbed384d
Stable URL of this page: https://metraipolis.com/documento/dc_c19f0436dab1

How to cite

Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials — arXiv:2610.09837 — arXiv (en), https://arxiv.org/abs/2610.09837 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c19f0436dab1 (Metraipolis archive dc_c19f0436dab1, sha256 75e0c311422846f25a696db3c048dec8ce5d8ef1ca7ead95864b1800cbed384d)

BibTeX

@misc{dcc19f0436dab1,
  title = {Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials — arXiv:2610.09837},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09837},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c19f0436dab1; sha256 75e0c311422846f25a696db3c048dec8ce5d8ef1ca7ead95864b1800cbed384d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09837. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Universal machine learning force-field generalization to multicomponent environments generated by compositional design remains insufficiently assessed. We construct a benchmark of 7,599 multicomponent configurations inspired by high-entropy design, elemental substitution and anion mixing. Eleven pretrained models are evaluated against density functional theory for energies, forces and stresses, with assessment extended to elastic, vibrational and adsorption-related properties. Force errors are analysed through training-reference coverage, local geometric heterogeneity, distance directionality and elemental response. Distances to training-reference environments reveal a qualitative association between coverage differences and increasing errors, while substantial variation remains at similar distances. Higher-error groups show greater local geometric heterogeneity, although OMat24 provides broad coverage of these environments. Relative to training-reference pair medians, errors remain low near the median, rise steeply on the compression side and increase more weakly on the extension side. After matching element pairs and absolute distance deviations, compression-side force errors are 1.81-1.95 times extension-side errors. Model-predicted pairwise interaction curves show greater curvature under compression. Fitting difficulty in independent elemental systems correlates with electronic band-energy responses to atomic displacements and Fermi-level shifts, and a similar pattern is observed in multicomponent systems. In parameter-matched comparisons, spherical-harmonic representations with maximum degrees of 2 and 4 lower test force errors for 38 and 40 of 43 elements, respectively, while differences in elemental difficulty remain. These findings inform force-field selection for experimental compositional design and identify targets for training-data sampling and model representations.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Materials Science (cond-mat.mtrl-sci) ; Machine Learning (cs.LG); Computational Physics (physics.comp-ph).
Pagina del lavoro: https://arxiv.org/abs/2610.09837 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T12:14:15.422Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).