Mastodawn

Christos Argyropoulos MD, PhD Sep 7, 2024

This review of #MCMC convergence diagnostics is really good

https://www.annualreviews.org/content/journals/10.1146/annurev-statistics-031219-041300

and pairs very well with this article discussing convergence of #MonteCarlo in #GPU friendly samplers
https://arxiv.org/abs/2110.13017

Only for the serious #Bayesians who refuse to do sloppy numerical work.
#statistics

Convergence Diagnostics for Markov Chain Monte Carlo | Annual Reviews

Markov chain Monte Carlo (MCMC) is one of the most useful approaches to scientific computing because of its flexible construction, ease of use, and generality. Indeed, MCMC is indispensable for performing Bayesian analysis. Two critical questions that MCMC practitioners need to address are where to start and when to stop the simulation. Although a great amount of research has gone into establishing convergence criteria and stopping rules with sound theoretical foundation, in practice, MCMC users often decide convergence by applying empirical diagnostic tools. This review article discusses the most widely used MCMC convergence diagnostic tools. Some recently proposed stopping rules with firm theoretical footing are also presented. The convergence diagnostics and stopping rules are illustrated using three detailed examples.

Christos Argyropoulos Sep 7, 2024

This review of #MCMC convergence diagnostics is really good

https://www.annualreviews.org/content/journals/10.1146/annurev-statistics-031219-041300

and pairs very well with this article discussing convergence of #MonteCarlo in #GPU friendly samplers
https://arxiv.org/abs/2110.13017

Only for the serious #Bayesians who refuse to do sloppy numerical work.
#statistics

Convergence Diagnostics for Markov Chain Monte Carlo | Annual Reviews

Markov chain Monte Carlo (MCMC) is one of the most useful approaches to scientific computing because of its flexible construction, ease of use, and generality. Indeed, MCMC is indispensable for performing Bayesian analysis. Two critical questions that MCMC practitioners need to address are where to start and when to stop the simulation. Although a great amount of research has gone into establishing convergence criteria and stopping rules with sound theoretical foundation, in practice, MCMC users often decide convergence by applying empirical diagnostic tools. This review article discusses the most widely used MCMC convergence diagnostic tools. Some recently proposed stopping rules with firm theoretical footing are also presented. The convergence diagnostics and stopping rules are illustrated using three detailed examples.

Junpeng Lao 🏂Feb 20, 2023

Looks like the program for #BayesComp2023 is out (bayescomp2023.com/)! I am very excited to visit Levi, #Finland in March! Super excited to:
- meeting fellow #Bayesians @avehtari, @sethaxen
- presenting short talk on functional programming for Bayesian workflow with #tensorflow_probability and #blackjax
- skiing some Finland powder

Berlin Bayesians Dec 30, 2022

Hi there, it is time for an #Introduction !

We are #Berlin #Bayesians, a group for Bayesian enthusiasts and practitioners based in Berlin. We organize talks (mostly in person, sometimes online) on all things Bayesian.

You can check out our events here: https://www.meetup.com/berlinbayesians/

If you'd like to speak at our meetup you can: present your research, give a tutorial on a Bayesian framework or summarize a paper. Feel free to write us, even if you are not based in Berlin.

Berlin Bayesians (Berlin, Germany)

We meet monthly to learn about Bayesian approaches to data analysis, machine learning, and statistics.Join our Slack workspace to talk to the community:https://join.slack.com/t/berlinbayesians/shared_

Meetup

Show thread

A. Jordan Nafa Nov 29, 2022

I'm at the point where I have to actually tackle this question for the dissertation (also still need to do a bunch of simulation for my BMAME approach), but am torn so pitching this to #statsodon and #bayesians in the hope that someone either has a better answer than I do or can tell me that I'm just wildly overthinking this 🤷‍♂️

Seth Axen 🪓

Nov 17, 2022

I'm very excited to be attending #BayesComp2023 in Levi, #Finland in March and looking forward to meeting any fellow #Bayesians who will also be there!

https://bayescomp2023.com/