Metraipolis › archive › dc_9ee2b1a995da

Kernel Reboot: Breaking the Boundaries of Neural Tangent Kernels for Neural Fields — arXiv:2609.03117

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.03117
License (as archived)
CC0-1.0
Archived
2026-10-17T02:07:46.563Z
SHA-256 of the archived text
83714d850cb87309463c2a677771e815141e2ab6f8922a43788b5e2b79ef2912
Stable URL of this page
https://metraipolis.com/documento/dc_9ee2b1a995da

How to cite

Kernel Reboot: Breaking the Boundaries of Neural Tangent Kernels for Neural Fields — arXiv:2609.03117 — arXiv (en), https://arxiv.org/abs/2609.03117 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9ee2b1a995da (Metraipolis archive dc_9ee2b1a995da, sha256 83714d850cb87309463c2a677771e815141e2ab6f8922a43788b5e2b79ef2912)
BibTeX
@misc{dc9ee2b1a995da,
  title = {Kernel Reboot: Breaking the Boundaries of Neural Tangent Kernels for Neural Fields — arXiv:2609.03117},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03117},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9ee2b1a995da; sha256 83714d850cb87309463c2a677771e815141e2ab6f8922a43788b5e2b79ef2912},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03117. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural fields (NFs) map continuous coordinates to signals such as color or density, but fast high-quality reconstruction from sparse observations remains difficult. Classical Neural Tangent Kernel (NTK) regression gives closed-form fits, yet it is fundamentally linear and cannot accumulate reusable task priors. We develop three algorithms that address these gaps. NTK-KIP learns a distilled support set of coordinates (and optional labels) so that a finite NTK can inpaint large missing regions from little observed data, yielding a compact non-linear representation instead of a raw kernel solve. MetaQuill meta-learns a shared initialization for an INR so that new scenes can be adapted by updating only a small task-specific weight offset, which provides true feature learning and a reusable prior. Finally, MetaQuill-KIP fuses both ideas: it seeds the task with a KIP-style non-linear warm start, then refines only that small offset around the meta-learned initialization. MetaQuill-KIP achieves high-PSNR reconstructions and semantically plausible inpainting under very sparse observations, while requiring only lightweight per-instance adaptation, whereas diffusion-style baselines typically depend on large pretrained generative priors and costly per-image tuning. This shows that NTK-driven neural fields can be made both non-linear and meta-learnable, narrowing the gap between analytic kernels and practical few-shot reconstruction.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.03117 — 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 2026-10-17T01:47:25.525Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).