Metraipolis › archive › dc_029cd308ec56

Conditional Mean Flows: Few-Step Densities for Simulation-Based Inference — arXiv:2610.21657

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.21657
License (as archived)
CC0-1.0
Archived
2026-10-16T10:25:14.951Z
SHA-256 of the archived text
526013c6f7a196e27df0def5850ca6d672cdbb1b8901fa6ff5bab0370e30fb95
Stable URL of this page
https://metraipolis.com/documento/dc_029cd308ec56

How to cite

Conditional Mean Flows: Few-Step Densities for Simulation-Based Inference — arXiv:2610.21657 — arXiv (en), https://arxiv.org/abs/2610.21657 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_029cd308ec56 (Metraipolis archive dc_029cd308ec56, sha256 526013c6f7a196e27df0def5850ca6d672cdbb1b8901fa6ff5bab0370e30fb95)
BibTeX
@misc{dc029cd308ec56,
  title = {Conditional Mean Flows: Few-Step Densities for Simulation-Based Inference — arXiv:2610.21657},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.21657},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_029cd308ec56; sha256 526013c6f7a196e27df0def5850ca6d672cdbb1b8901fa6ff5bab0370e30fb95},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21657. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Flow matching has emerged as a powerful generative family for simulation-based inference (SBI). However, sampling and density evaluation require numerical integration at inference time. This makes posterior sampling in neural posterior estimation costly. More importantly, it can make repeated likelihood evaluation prohibitively expensive and impractical for Markov chain Monte Carlo (MCMC) sampling, which may require thousands of repeated density evaluations. Here, we develop conditional mean flows for posterior and likelihood approximation, combining few-step sampling with tractable density evaluation. We show that the mean flow objective identifies conditional flow maps and that independent masking enables direct estimation of marginal likelihoods. We evaluate both posterior and likelihood estimation across several tasks and show that conditional mean flows retain the flexibility of flow-based density estimators while significantly reducing inference time. In particular, single-step conditional mean flow density evaluation yields estimation error comparable to 200-step flow matching while enabling about 200-fold faster downstream MCMC sampling.

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).
Pagina del lavoro: https://arxiv.org/abs/2610.21657 — 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 2026-10-16T08:48:02.913Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).