Metraipolis › archive › dc_e303bdb84bf8

Multilevel Gaussian Processes with Applications to Inverse Problems — arXiv:2610.23202

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.23202
License (as archived)
CC0-1.0
Archived
2026-10-19T09:38:31.604Z
SHA-256 of the archived text
77731059f8019bb78138d0a29da4b147ec461aca0b1ba8855cae57b4b64280f7
Stable URL of this page
https://metraipolis.com/documento/dc_e303bdb84bf8

How to cite

Multilevel Gaussian Processes with Applications to Inverse Problems — arXiv:2610.23202 — arXiv (en), https://arxiv.org/abs/2610.23202 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e303bdb84bf8 (Metraipolis archive dc_e303bdb84bf8, sha256 77731059f8019bb78138d0a29da4b147ec461aca0b1ba8855cae57b4b64280f7)
BibTeX
@misc{dce303bdb84bf8,
  title = {Multilevel Gaussian Processes with Applications to Inverse Problems — arXiv:2610.23202},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23202},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e303bdb84bf8; sha256 77731059f8019bb78138d0a29da4b147ec461aca0b1ba8855cae57b4b64280f7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23202. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present a multilevel Gaussian process (GP) regression framework for approximating functions that are accessible only through a hierarchy of approximations of increasing accuracy and computational cost. One example is the response surface of a parametrized partial differential equation, where the hierarchy arises from the resolution of the numerical solver. Modeling level increments independently with Matérn-kernel GPs and allocating samples flexibly across discretization levels yields substantial computational savings over single-level GP regression. We derive $L^2$ convergence rates for the posterior mean and variance, accounting for maximum likelihood estimation of the marginal variance and for kernel misspecification. When increment regularity or solver cost is unknown a priori, an adaptive sequential-design algorithm with cost-adapted active learning criteria allocates samples across levels. We apply the surrogate to Bayesian inverse problems, for which Hellinger distance bounds between the true and approximate posteriors are known. Numerical experiments illustrate the theoretical rates and the resulting computational savings.

Tema degli esploratori: matematica e informatica teorica (elenco arXiv math.NA, nuovi lavori del giorno).
Categorie arXiv: Numerical Analysis (math.NA).
Pagina del lavoro: https://arxiv.org/abs/2610.23202 — 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 2026-10-19T09:34:29.793Z dalla pagina https://arxiv.org/list/math.NA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).