Metraipolis › archive › dc_273fcf26faad

Neural Fields Encode Adaptation Geometry — arXiv:2610.07253

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.07253
License (as archived): CC0-1.0
Archived: 2026-10-07T11:40:59.399Z
SHA-256 of the archived text: 4b23ba23b0042e9fca1db390abc5fb6f714e436c5820e0b0a1c6599b6ace1a9d
Stable URL of this page: https://metraipolis.com/documento/dc_273fcf26faad

How to cite

Neural Fields Encode Adaptation Geometry — arXiv:2610.07253 — arXiv (en), https://arxiv.org/abs/2610.07253 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_273fcf26faad (Metraipolis archive dc_273fcf26faad, sha256 4b23ba23b0042e9fca1db390abc5fb6f714e436c5820e0b0a1c6599b6ace1a9d)

BibTeX

@misc{dc273fcf26faad,
  title = {Neural Fields Encode Adaptation Geometry — arXiv:2610.07253},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07253},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_273fcf26faad; sha256 4b23ba23b0042e9fca1db390abc5fb6f714e436c5820e0b0a1c6599b6ace1a9d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07253. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural fields are usually evaluated by how well they reconstruct an observation. We show that this misses two useful properties of a fitted network: how easily it can adapt to new observations, and what its weights retain from earlier ones. We study these properties as adaptation geometry. For images, we meta-learn class-specific initializations, adapt each one to a new image, and measure how much the network must change to fit it. A simple local linear model closely predicts this adaptation cost, while replacing one network's tangent kernel with another's substantially worsens the prediction. Adaptation thus depends on the local geometry of the fitted network, not only on its current reconstruction. For physical fields, we repeatedly fit the same network to observations from a sequence. Its weights then retain information about that history. When two wave histories end at exactly the same observation, the final weights recover the sign of the wave velocity with 68.6% accuracy, whereas the current observation alone contains no such information and gives 50%. These two phenomena are quantitatively linked: tangent-kernel eigenvalues predict both which changes are easy to learn and how quickly they are overwritten by later fitting. Together, these results show that neural fields contain useful information beyond what they currently reconstruct: in how they can change and in how they got there.

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/2610.07253 — 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 2026-10-07T11:19:24.741Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).