Metraipolis › archive › dc_359790507bed

Learn-Then-Differentiate Gradient Estimation — arXiv:2609.38842

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.38842
License (as archived)
CC0-1.0
Archived
2026-10-07T13:47:22.461Z
SHA-256 of the archived text
c2168222939b97a6431b864912b3ee5ab7663588d2f21c1c7ab70699999b6648
Stable URL of this page
https://metraipolis.com/documento/dc_359790507bed

How to cite

Learn-Then-Differentiate Gradient Estimation — arXiv:2609.38842 — arXiv (en), https://arxiv.org/abs/2609.38842 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_359790507bed (Metraipolis archive dc_359790507bed, sha256 c2168222939b97a6431b864912b3ee5ab7663588d2f21c1c7ab70699999b6648)
BibTeX
@misc{dc359790507bed,
  title = {Learn-Then-Differentiate Gradient Estimation — arXiv:2609.38842},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38842},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_359790507bed; sha256 c2168222939b97a6431b864912b3ee5ab7663588d2f21c1c7ab70699999b6648},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38842. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learn-then-differentiate (LTD) estimates gradients by fitting a model to simulation outputs and differentiating it. We develop a unified framework explaining what LTD differentiates and how accurately it estimates gradients. For models with a weighted representation, LTD differentiates a learned representation of the underlying probability measure. We then show how accuracy guarantees for fitted models translate into guarantees for gradients and higher-order derivatives, with rates approaching the standard Monte Carlo rate under suitable smoothness conditions. The framework recovers established results for kernel regression, local polynomial regression, and kernel ridge regression, and yields further guarantees for multiple kernel learning and smooth neural networks. These results provide a common foundation for understanding and analyzing LTD across learning methods.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.LG).
Categorie arXiv: Computation (stat.CO) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.38842 — 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 2026-10-07T13:18:10.598Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).