Metraipolis › archive › dc_5678ff08748e
MCMC Can See Can't See — arXiv:2610.20020
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.20020
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T14:24:23.589Z
- SHA-256 of the archived text
bfddaa09771531505db56917994a7bdbf183c177f72df1d98b10949185504f43- Stable URL of this page
https://metraipolis.com/documento/dc_5678ff08748e
How to cite
MCMC Can See Can't See — arXiv:2610.20020 — arXiv (en), https://arxiv.org/abs/2610.20020 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5678ff08748e (Metraipolis archive dc_5678ff08748e, sha256 bfddaa09771531505db56917994a7bdbf183c177f72df1d98b10949185504f43)
BibTeX
@misc{dc5678ff08748e,
title = {MCMC Can See Can't See — arXiv:2610.20020},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.20020},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5678ff08748e; sha256 bfddaa09771531505db56917994a7bdbf183c177f72df1d98b10949185504f43},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20020. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
It is folklore that passing "convergence" diagnostics does not guarantee that a Markov chain has converged. We prove this as an impossibility for any finite checklist of diagnostics. Running more chains, longer chains, and smarter diagnostics does not avoid this impossibility in certifying convergence. Under regularity assumptions, we determine how the worst-case undetected non-convergence decays as more diagnostics are added, and how to choose diagnostics optimally. Folklore suggests that the sampler's slowest mixing modes are the "dangerous" ones that must be monitored; we prove that the slow modes yield nearly optimal diagnostics as sampling proceeds. Given a source class, we also give an anytime-valid bound on the remaining discrepancy and show how undetected bias caps the effective sample size that a finite checklist can certify. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv stat.ML, nuovi lavori del giorno). Categorie arXiv: Computation (stat.CO) ; Statistics Theory (math.ST); Machine Learning (stat.ML). Pagina del lavoro: https://arxiv.org/abs/2610.20020 — 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 2026-10-15T13:54:20.321Z dalla pagina https://arxiv.org/list/stat.ML/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).