Metraipolis › archive › dc_af20090d642a

A budget-dependent crossover between coverage- and response-based training-set selection for machine-learned interatomic potentials — arXiv:2609.05877

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.05877
License (as archived)
CC0-1.0
Archived
2026-10-16T13:56:24.116Z
SHA-256 of the archived text
58795b98a33ef9b0058e2ac048218040a158b579fe4b3859c619a73cf6474679
Stable URL of this page
https://metraipolis.com/documento/dc_af20090d642a

How to cite

A budget-dependent crossover between coverage- and response-based training-set selection for machine-learned interatomic potentials — arXiv:2609.05877 — arXiv (en), https://arxiv.org/abs/2609.05877 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_af20090d642a (Metraipolis archive dc_af20090d642a, sha256 58795b98a33ef9b0058e2ac048218040a158b579fe4b3859c619a73cf6474679)
BibTeX
@misc{dcaf20090d642a,
  title = {A budget-dependent crossover between coverage- and response-based training-set selection for machine-learned interatomic potentials — arXiv:2609.05877},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.05877},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_af20090d642a; sha256 58795b98a33ef9b0058e2ac048218040a158b579fe4b3859c619a73cf6474679},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.05877. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Selecting compact training sets for machine-learned interatomic potentials requires deciding whether to preserve structural diversity or target configurations on which models disagree. The better choice can depend on how much data is retained, making a comparison at one training-set size insufficient. Here we link selection criteria to prediction accuracy through a budget-resolved comparison of retrained MACE models on GAP-20 Carbon and pooled revised MD17. Structural coverage is compared with a response-guided selector that targets disagreement between a coverage-trained model and a full-data reference. This retrospective response witness tests the value of model disagreement for compressing an already labelled pool. At 5\%, coverage gives smaller absolute deviations from the full-data error than random sampling across four force endpoints in both datasets. The witness has larger deviations than coverage at 1\% and 5\%, but the ordering reverses at 20\%. At 20\%, witness-selected models also lower direct held-out force errors by 0.46--5.89\% relative to coverage, with all eight paired training-seed intervals favouring the witness. Six errors fall below the full-data reference. Mean force-error reductions are 0.164--0.167~meV~$\textÅ^{-1}$, with larger gains for tail and masked endpoints. Complementary analyses show that learned similarity preserves the coverage ranking, while selecting by frozen-model error gives higher error than embedding coverage. These findings establish retained-data budget as a deciding variable in atomistic training-set selection and provide a direct test of when response-guided compression improves on structural coverage.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.05877 — 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 2026-10-16T13:39:42.818Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).