[ABE-L] Ciclo de Palestras - PPGE-IM/UFRJ - 15 de Outubro

[Maria Eulalia Vares]

Prezados colegas,

A nossa próxima palestra ocorrerá na quarta-feira, 15 de OUTUBRO, no
horário das 15h30 às 17h00, Local: Laboratório de Sistemas Estocásticos
(LSE), Sala I-044-B, Centro de Tecnologia - UFRJ.

*Palestrante: *Flávio Bambirra (UFMG)

*Título*:Scalable Bernoulli Factory MCMC for Intractable Marginalised
Posteriors
*Resumo*: Bernoulli factory MCMC algorithms implement accept-reject Markov
chains without explicit computation of acceptance probabilities, and are
used to target posterior distributions associated with intractable
likelihood models. Intractable likelihoods naturally arise in
continuous-time models and mixture distributions, or from the
marginalisation of a tractable augmented model. Bernoulli factory MCMC
algorithms often mix better than alternatives that target a tractable
augmented posterior. However, for a likelihood that factorizes over
observations, we show that their computational performance typically
deteriorates exponentially with data size. To address this, we propose a
simple divide-and-conquer Bernoulli factory MCMC algorithm and prove that
it has polynomial complexity of degree between 1 and 2, with the exact
degree depending on the existence of efficient unbiased estimators of the
intractable likelihood ratio. We demonstrate the effectiveness of our
approach with applications to Bayesian inference in two intractable
likelihood models, and observe respective polynomial cost of degree 1.2 and
1 in the data size.

Mais informações:
https://ppge.im.ufrj.br/ciclo-de-palestras-segundo-semestre-de-2025/

Organizadores: Maria Eulalia Vares e Widemberg S Nobre

Atenciosamente,


Maria Eulalia Vares
Professora Titular - Instituto de Matemática - UFRJ
Coordenadora do Programa de Pós-Graduação em Estatística
https://ppge.im.ufrj.br/
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://lists.ime.usp.br/pipermail/abe/attachments/20251013/fbefcec5/attachment.htm>