Metraipolis › archive › dc_e8322cf08d5e

Resolution Allocation with Learnable Resizing and Spectral Filter Parameterization — arXiv:2610.20274

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20274
License (as archived)
CC0-1.0
Archived
2026-10-15T14:19:58.161Z
SHA-256 of the archived text
3a3fe230136510518932806f1c51b171df478c7fd5a6177c2b853c263cb9f231
Stable URL of this page
https://metraipolis.com/documento/dc_e8322cf08d5e

How to cite

Resolution Allocation with Learnable Resizing and Spectral Filter Parameterization — arXiv:2610.20274 — arXiv (en), https://arxiv.org/abs/2610.20274 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e8322cf08d5e (Metraipolis archive dc_e8322cf08d5e, sha256 3a3fe230136510518932806f1c51b171df478c7fd5a6177c2b853c263cb9f231)
BibTeX
@misc{dce8322cf08d5e,
  title = {Resolution Allocation with Learnable Resizing and Spectral Filter Parameterization — arXiv:2610.20274},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20274},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e8322cf08d5e; sha256 3a3fe230136510518932806f1c51b171df478c7fd5a6177c2b853c263cb9f231},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20274. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learnable downsampling and spectral-domain filter parameterizations have each made convolutional networks more flexible, but their interaction has not been systematically examined. We bridge this gap by examining convolutional kernels learned in the spatial and spectral domains, together with learned downsampling ratios under a cropped resizing implementation. We find that spectral and spatial kernels respond differently to learned resolution allocation, with larger gains for the spectral kernel. We observe this consistently on three datasets (EuroSAT, CIFAR-10 and STL-10) and at two model sizes (differences in accuracy gain of +0.013 to +0.028 on held-out test sets). Part of this difference reflects the greater gain of the spectral kernel from retaining more spatial area, but the difference remains positive when total retained area is matched. This pattern is observed across datasets and model sizes, although its magnitude and dependence on the complexity budget vary with model size and the resizing implementation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.20274 — 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 2026-10-15T13:53:35.721Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).