Metraipolis › archive › dc_f5455bea676d

Susceptibilities for Neural Networks Learning from Physical Data — arXiv:2611.02632

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02632
License (as archived)
CC0-1.0
Archived
2026-11-03T11:22:49.720Z
SHA-256 of the archived text
affc57d0ffe099845749adb8facf616556a8daf30c41b0228636e2b19ccf6eec
Stable URL of this page
https://metraipolis.com/documento/dc_f5455bea676d

How to cite

Susceptibilities for Neural Networks Learning from Physical Data — arXiv:2611.02632 — arXiv (en), https://arxiv.org/abs/2611.02632 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f5455bea676d (Metraipolis archive dc_f5455bea676d, sha256 affc57d0ffe099845749adb8facf616556a8daf30c41b0228636e2b19ccf6eec)
BibTeX
@misc{dcf5455bea676d,
  title = {Susceptibilities for Neural Networks Learning from Physical Data — arXiv:2611.02632},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02632},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f5455bea676d; sha256 affc57d0ffe099845749adb8facf616556a8daf30c41b0228636e2b19ccf6eec},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02632. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural networks that are trained on data arising from a physical system must somehow learn regularities induced by the underlying physical laws. In this setting, concepts from statistical physics provide powerful tools for analysing how physical laws influence the learning process. In this paper we explain how susceptibilities --- which measure the response of a neural network to perturbations of a parameter of the data distribution --- can be applied when learning from physical data. We show that such susceptibilities can identify the conditions under which specific input features are most informative for learning, guiding data selection, a prediction we validate experimentally. For distributions lacking a natural parameter, we propose introducing one via a physically meaningful scalar function of the physical state, and demonstrate that susceptibilities track the emergence of specific capabilities (such as collision detection) during training.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.02632 — 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 2026-11-03T11:02:37.630Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).