Metraipolis › archive › dc_3a57e549d72d

From Global to Tail Convergence: Dynamics of $α$-Stable Score-Based Generative Models — arXiv:2610.19773

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19773
License (as archived)
CC0-1.0
Archived
2026-10-15T14:17:46.490Z
SHA-256 of the archived text
850ab0ce7e68601cead87741831d9e5dc0ea19cc1f463ff93bcfef2199d69d73
Stable URL of this page
https://metraipolis.com/documento/dc_3a57e549d72d

How to cite

From Global to Tail Convergence: Dynamics of $α$-Stable Score-Based Generative Models — arXiv:2610.19773 — arXiv (en), https://arxiv.org/abs/2610.19773 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3a57e549d72d (Metraipolis archive dc_3a57e549d72d, sha256 850ab0ce7e68601cead87741831d9e5dc0ea19cc1f463ff93bcfef2199d69d73)
BibTeX
@misc{dc3a57e549d72d,
  title = {From Global to Tail Convergence: Dynamics of $α$-Stable Score-Based Generative Models — arXiv:2610.19773},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19773},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3a57e549d72d; sha256 850ab0ce7e68601cead87741831d9e5dc0ea19cc1f463ff93bcfef2199d69d73},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19773. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Score-based generative models driven by $\alpha$-stable L{é}vy processes have been proposed as an alternative to Brownian diffusion models, with empirical evidence suggesting improved sampling and better modeling of heavy-tailed data. Yet theoretical guarantees for these models remain limited. We develop a non-asymptotic convergence analysis for scale-preserving $\alpha$-stable generative dynamics, covering both compactly supported and power-law targets. In p-Wasserstein distance, we quantify the effects of early stopping, initialization, score approximation, and time discretization, and show exponential decay of the initialization error. A comparison with Brownian dynamics reveals that global Wasserstein convergence alone does not fundamentally distinguish the two models: both initialization errors decay exponentially with the forward horizon on the common class of heavy-tailed targets. A sharp separation emerges instead in the tails. At any positive generation time, stable dynamics produces power-law tails with index $\alpha$, whereas Brownian dynamics remains light-tailed. For a target with tail index $\nu$, asymptotically correct extreme-quantile scaling is possible precisely when $\alpha$=$\nu$. These results identify tail-index matching as a principled mechanism behind the ability of stable generative models to capture extreme events.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Probability (math.PR); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.19773 — 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 2026-10-15T13:53:35.640Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).