[ABE-L] Seminarios em High Dimensional Data Analysis

[Ronaldo Dias]

Prezados(as),

Dando sequência aos nosso seminários em <AD,AD> teremos Victor F. Souza
(IMECC-Unicamp), apresentando

Título: Bayesian analysis of Markov random fields with intractable constants
Resumo: Most Markov random field models have intractable normalizing
constants making direct likelihood evaluation not possible. While parameter
estimation in a classical context can be done without evaluating the
likelihood function, Bayesian methods for posterior sampling are unable to
avoid it with the same strategies. A quick introduction to Markov random
field models is presented, followed by the general idea behind estimation
methods and a version of Metropolis-Hastings algorithm proposed in Atchadé,
Y. F., Lartillot, N., & Robert, C. (2013). We show a simulation study to
measure the algorithm performance on a simple Gaussian scenario, where the
posterior distribution can be obtained exactly, and in the Ising model.

Aos que puderem comparecer, sejam bem vindos

Data: 17/04/19
Local: sala 221. IMECC-Unicamp
Hora: 13:00hs.

-- 
Ronaldo Dias, Ph.D.
Professor
Dept. of Statistics-IMECC, UNICAMP
www.ime.unicamp.br/~dias
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <https://lists.ime.usp.br/archives/abe/attachments/20190415/cd9849ca/attachment.html>