Metraipolis › archive › dc_05c6aedc4da8
Towards Out-of-Distribution Uncertainty Quantification for Machine-Learned Interatomic Potentials via Pre-Training — arXiv:2610.41801
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.41801
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T05:40:58.324Z
- SHA-256 of the archived text
4aa5eb25816eac5ea2ef8faa9ae599b279e21d19fc55e4a10b1c7135ac165f1d- Stable URL of this page
https://metraipolis.com/documento/dc_05c6aedc4da8
How to cite
Towards Out-of-Distribution Uncertainty Quantification for Machine-Learned Interatomic Potentials via Pre-Training — arXiv:2610.41801 — arXiv (en), https://arxiv.org/abs/2610.41801 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_05c6aedc4da8 (Metraipolis archive dc_05c6aedc4da8, sha256 4aa5eb25816eac5ea2ef8faa9ae599b279e21d19fc55e4a10b1c7135ac165f1d)
BibTeX
@misc{dc05c6aedc4da8,
title = {Towards Out-of-Distribution Uncertainty Quantification for Machine-Learned Interatomic Potentials via Pre-Training — arXiv:2610.41801},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.41801},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_05c6aedc4da8; sha256 4aa5eb25816eac5ea2ef8faa9ae599b279e21d19fc55e4a10b1c7135ac165f1d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41801. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Informative uncertainty quantification (UQ) is essential for the practical deployment of machine-learned interatomic potentials (MLIPs), as it provides insight into their reliability across chemical space. Although numerous post-hoc uncertainty estimation methods for MLIPs have recently been proposed, most have been trained in low-data regimes, restricting their applicability to narrow domains of interest. Inspired by the emergence of foundational MLIPs, we propose pre-training uncertainty estimators on large, diverse datasets to enable broader applicability. We pre-train several UQ models on large materials science datasets to quantify the uncertainty of different MLIPs, and evaluate their performance on both in-distribution and out-of-domain (OOD) adsorption-energy predictions. We find that pre-trained uncertainty estimators provide viable zero-shot uncertainty estimates across multiple datasets, while fine-tuning consistently yields positive correlations between predicted uncertainty and prediction error, even on the most challenging datasets. These results lay the groundwork for foundational UQ models for MLIPs, ultimately facilitating their deployment in practical settings by providing users with a priori information about model reliability and applicability. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Chemical Physics (physics.chem-ph). Pagina del lavoro: https://arxiv.org/abs/2610.41801 — 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 2026-11-03T05:26:25.715Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).