Metraipolis › archive › dc_d143ff8df907

Point interpolation and distributional universality in flow-based generative models — arXiv:2611.04207

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04207
License (as archived)
CC0-1.0
Archived
2026-11-04T18:55:50.859Z
SHA-256 of the archived text
37f7ee81bbb144a8cb79c1c8294a9f8bc6d8551ff17aa1067283578e9de4a28f
Stable URL of this page
https://metraipolis.com/documento/dc_d143ff8df907

How to cite

Point interpolation and distributional universality in flow-based generative models — arXiv:2611.04207 — arXiv (en), https://arxiv.org/abs/2611.04207 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d143ff8df907 (Metraipolis archive dc_d143ff8df907, sha256 37f7ee81bbb144a8cb79c1c8294a9f8bc6d8551ff17aa1067283578e9de4a28f)
BibTeX
@misc{dcd143ff8df907,
  title = {Point interpolation and distributional universality in flow-based generative models — arXiv:2611.04207},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04207},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d143ff8df907; sha256 37f7ee81bbb144a8cb79c1c8294a9f8bc6d8551ff17aa1067283578e9de4a28f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04207. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Motivated by the expressive power of neural ordinary differential equations in flow-based generative modeling, we study the approximation of probability measures by pushforwards generated by parameterized families of Lipschitz vector fields. For Wasserstein distances, we establish a quantitative relationship between finite-sample point interpolation and distributional approximation, and derive universality criteria based on $L^p$-approximation of transport maps. In particular, affine-invariant families containing a non-affine vector field are universal from compactly supported absolutely continuous base measures. For total variation, we show that coordinate-wise dynamics are insufficient in dimensions $d\geq 2$ and establish sufficient conditions based on one-dimensional Sobolev approximation and shear-type mixing. We further examine the role of point interpolation, establishing when different regimes relate to one another and when approximate interpolation can be made exact.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG); Optimization and Control (math.OC).
Pagina del lavoro: https://arxiv.org/abs/2611.04207 — 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 2026-11-04T18:35:37.522Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).