Metraipolis › archive › dc_cb2ec4784b45

Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the sphere — arXiv:2610.23821

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.23821
License (as archived)
CC0-1.0
Archived
2026-10-19T09:21:34.349Z
SHA-256 of the archived text
e910645234786848bef72b6a6aac649f4a6ab769f234932852579864208fdb83
Stable URL of this page
https://metraipolis.com/documento/dc_cb2ec4784b45

How to cite

Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the sphere — arXiv:2610.23821 — arXiv (en), https://arxiv.org/abs/2610.23821 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cb2ec4784b45 (Metraipolis archive dc_cb2ec4784b45, sha256 e910645234786848bef72b6a6aac649f4a6ab769f234932852579864208fdb83)
BibTeX
@misc{dccb2ec4784b45,
  title = {Learning without relearning physical symmetry: an SO(3)-equivariant convolutional network (CNN) on the sphere — arXiv:2610.23821},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23821},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cb2ec4784b45; sha256 e910645234786848bef72b6a6aac649f4a6ab769f234932852579864208fdb83},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23821. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Many of the most important datasets in the physical sciences live natively on the sphere, and these data often carry no preferred position or orientation. Using methods that do not reflect this symmetry leads to far higher simulation requirements, which can be catastrophic for scientific analyses (forcing reliance on cheap, imperfect models that bias results). We present a novel spherical implementation of a gauge equivariant mesh architecture that is specifically designed for the Hierarchical Equal Area isoLatitude Pixelization (\healpix{}) of the sphere. The $\mathrm{SO}(3)$-equivariant convolution is built from steerable filters, parallel transport and spin-weighted features. The architecture supplies the exact symmetry, so the network does not spend training data learning it. The convolution operations are constructed to give provably exact equivariance, given a band-limit. Our full CNNs can achieve SO(3)-invariance error at per-cent-level and lower, which is controlled through well-understood hyperparameters. On benchmarks, our technique gives the most accurate dense gradient regression and competitive cosmic microwave background foreground removal and denoising. The improved simulation efficiency matters most in simulation-based inference (SBI), where high-fidelity simulations are expensive and neural compressors require large training sets, so a fixed budget forces analyses with less accurate simulators. In cosmology lensing SBI, our technique yields the tightest posteriors at matched simulation count. This approach is a novel method that will enable new convolutional neural network analyses, including generative and inference tasks, for processes and data on the sphere.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (stat.ML) ; Cosmology and Nongalactic Astrophysics (astro-ph.CO); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.23821 — 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 2026-10-19T09:17:41.735Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).