Metraipolis › archive › dc_bbb652a02473

Generator-Guided Inverse Sampling for Lévy-Driven Generative Models — arXiv:2608.10384

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.10384
License (as archived): CC0-1.0
Archived: 2026-10-19T06:19:20.485Z
SHA-256 of the archived text: c495f16ac1b59a6f4f48c52e810dbd853e76bd23c673c26e3b226cda735bb3a7
Stable URL of this page: https://metraipolis.com/documento/dc_bbb652a02473

How to cite

Generator-Guided Inverse Sampling for Lévy-Driven Generative Models — arXiv:2608.10384 — arXiv (en), https://arxiv.org/abs/2608.10384 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bbb652a02473 (Metraipolis archive dc_bbb652a02473, sha256 c495f16ac1b59a6f4f48c52e810dbd853e76bd23c673c26e3b226cda735bb3a7)

BibTeX

@misc{dcbbb652a02473,
  title = {Generator-Guided Inverse Sampling for Lévy-Driven Generative Models — arXiv:2608.10384},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10384},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bbb652a02473; sha256 c495f16ac1b59a6f4f48c52e810dbd853e76bd23c673c26e3b226cda735bb3a7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10384. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This paper studies inverse sampling for Lévy-driven generative models from the perspective of Markov generators. Unlike conventional diffusion models, Lévy-driven dynamics involve infinite jump activities, which makes their reverse process nonlocal and difficult to characterize using score information alone. We address this challenge by analyzing the forward and reversed generators. It is derived that the reversed jump component generally becomes a state-dependent Markov jump process governed by a nonlocal density ratio. This observation motivates a structured reverse sampler that decomposes the dynamics into diffusion, small jump, and large jump components. Based on this characterization, we develop a computationally tractable sampler for a class of isotropic linear Lévy SDEs with symmetric $\alpha$-stable jump components. For the jump component, the neural network is used only to amortize the rate of large jump activities, while jump amplitudes are generated from analytically derived conditional distributions, which improves interpretability and controllability. Efficient implementation techniques are further introduced under this setting to avoid expensive high-dimensional integration and sampling. The sampler is further adapted to approximate observation-guided sampling and applied to OFDM-SISO channel estimation under mixed Gaussian and impulsive noise. Simulations show robust estimation performance with a favorable tradeoff between complexity and performance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-12 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.10384 — 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 2026-10-19T06:11:45.079Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).