Metraipolis › archive › dc_fdf2026fb68f

Physics-Aligned Electronic Ground-State Learning Improves Generalization — arXiv:2610.10298

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10298
License (as archived)
CC0-1.0
Archived
2026-10-08T12:36:14.129Z
SHA-256 of the archived text
c61d31b20c6f38a023f43182455fced702e30e843867b1f70423e7996d01c009
Stable URL of this page
https://metraipolis.com/documento/dc_fdf2026fb68f

How to cite

Physics-Aligned Electronic Ground-State Learning Improves Generalization — arXiv:2610.10298 — arXiv (en), https://arxiv.org/abs/2610.10298 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fdf2026fb68f (Metraipolis archive dc_fdf2026fb68f, sha256 c61d31b20c6f38a023f43182455fced702e30e843867b1f70423e7996d01c009)
BibTeX
@misc{dcfdf2026fb68f,
  title = {Physics-Aligned Electronic Ground-State Learning Improves Generalization — arXiv:2610.10298},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10298},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fdf2026fb68f; sha256 c61d31b20c6f38a023f43182455fced702e30e843867b1f70423e7996d01c009},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10298. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine-learned interatomic potentials (MLIPs) excel at in-distribution tasks, accelerating drug and material development, yet they struggle to generalize out-of-distribution. We propose to push the cost-accuracy Pareto frontier by designing observable-agnostic electronic ground-state descriptor models (GSMs) with computational costs situated between MLIPs and Kohn-Sham density functional theory (KS-DFT). We align the learning objectives and architectures of GSMs with the governing equations of KS-DFT by enforcing physical constraints and removing optimization pressure on unphysical or irrelevant degrees of freedom. In our size-extrapolation experiments from QM9 to QM40, our combined contributions OrthoNormal-Loss (ON-Loss) and Grassmann Restricted Occupied-Orbital Training (GROOT) reach a 79.1% energy and 83.4% force mean absolute error (MAE) reduction over previous state-of-the-art density GSMs. For Hamiltonian GSMs, ON-Loss and Residual Optimal-gauge Conditioning-aware KS-Eq. Training (ROCKET) together reduce the energy and force MAEs of the strongest baseline by 99.8% and 95.9%, respectively. Using a self-consistency rejection criterion, we filter out extrapolation errors on QMugs, rejecting fewer than 0.4% of predictions while reaching an energy MAE of 0.07 mHa. Finally, we demonstrate the efficiency of label-free self-consistency fine-tuning, and transfer GSMs to reactive chemistry in Transition1x, reaching energy errors below chemical accuracy.

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); Quantum Physics (quant-ph).
Pagina del lavoro: https://arxiv.org/abs/2610.10298 — 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 2026-10-08T12:14:15.322Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).