Archive for Athens

Thucydides 2.0

Posted in Statistics with tags , , , , , , , , , , , , , , on May 24, 2026 by xi'an

Last weekend, I was listening to one of my favourite (France Inter) radio shows, Quand les dieux rôdaient sur la Terre (when gods roamed the Earth), and the story was about the siege of the tiny island of Melos by the Athenians and the subsequent massacre, based on the report given in Thucydides’ History of the Peloponnesian War and in particular the Melian Dialogue. I was unaware of this episode, but the modern tone of the dialogue excerpts was striking and made me think they equally applied to modern leaders… To wit,

“The strong do what they can, and the weak suffer what they must.” — Melian Dialogue, Book V

“Most people, in fact, will not take the trouble in finding out the truth, but are much more inclined to accept the first story they hear.” — Book I, §20

“Men naturally despise those who court them, but respect those who do not give way to them.” — Book III, Cleon’s speech

“It is a common mistake in going to war to begin at the wrong end — to act first, and wait for disaster to find out what to do.” — Book I, §78

“Words had to change their ordinary meaning and to take that which was now given them. Reckless audacity came to be considered the courage of a loyal ally; prudent hesitation, specious cowardice.” — Book III, §8

O’Bayes 2025 της Αθήνας

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , , on January 18, 2025 by xi'an

A reminder that the 2025 Objective Bayes (O’Bayes) Methodology Conference will take place in Athens, Greece, from June 8 to 12, 2025, which I will alas miss (for attending BayesComp the week after). Registration for the O’Bayes25 meeting is now open, with an increase on the 470€ fees in March. Speakers and discussants are available on the website and poster can be proposed till 28 February 28. The tutorials on Sunday will be given by Guido Consonni, Rianna de Heide, and Mark Stell (University of Warwick).

O’Bayes 2025 [με αθηναϊκό άρωμα]

Posted in Statistics with tags , , , , , , , , , , , , , , , , , , , on November 6, 2024 by xi'an


The next O’Bayes conference will take place in Athens, 8-12 June 2025. Registration is open, as well as a call for posters. As in (all?) earlier episodes of these Objective Bayes methodology conferences, speakers and discussants are invited by the scientific committee. And the first afternoon consists of tutorials on different aspects of Objective Bayes methodology. The conference will take place in the Stavros Niarchos Foundation Cultural Centre. (Unfortunately, I will not be able to attend O’Bayes 2025, once again, due to my attending BayesComp 2025 the week after.)

control variates [seminar]

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , , , , , , on November 5, 2021 by xi'an

Today, Petros Dellaportas (whom I have know since the early days of MCMC, when we met in CIRM) gave a seminar at the Warwick algorithm seminar on control variates for MCMC, reminding me of his 2012 JRSS paper. Based on the Poisson equation and using a second control variate to stabilise the Monte Carlo approximation do the first control variate. The difference with usual control variates is finding a first approximate G(x)-q(y|x)G(Y) to F-πF. And the first Poisson equation is using α(x,y)q(y|x) rather than π. Then the second expands log α(x,y)q(y|x) to achieve a manageable term.

Abstract: We provide a general methodology to construct control variates for any discrete time random walk Metropolis and Metropolis-adjusted Langevin algorithm Markov chains that can achieve, in a post-processing manner and with a negligible additional computational cost, impressive variance reduction when compared to the standard MCMC ergodic averages. Our proposed estimators are based on an approximate solution of the Poisson equation for a multivariate Gaussian target densities of any dimension.

I wonder if there were a neural network version that would first build G from scratch and later optimise it towards solving the Poisson equation. As in this recent arXival I haven’t read (yet).

MCMC with control variates

Posted in Books, Statistics, University life with tags , , , , , , , , , , on February 17, 2012 by xi'an

In the latest issue of JRSS Series B (74(1), Jan, 2012), I just noticed that no paper is “from my time” as co-editor, i.e. that all of them have been submitted after I completed my term in Jan. 2010. Given the two year delay, this is not that surprising, but it also means I can make comments on some papers w/o reservation! A paper I had seen earlier (as a reader, not as an editor nor as a referee!) is Petros Dellaportas’ and Ioannis Kontoyiannis’ Control variates for estimation based on  reversible Markov chain Monte Carlo samplers. The idea is one of post-processing MCMC output, by stabilising the empirical average via control variates. There are two difficulties, one in finding control variates, i.e. functions $\Psi(\cdot)$ with zero expectation under the target distribution, and another one in estimating the optimal coefficient in a consistent way. The paper solves the first difficulty by using the Poisson equation, namely that G(x)-KG(x) has zero expectation under the stationary distribution associated with the Markov kernel K. Therefore, if KG can be computed in closed form, this is a generic control variate taking advantage of the MCMC algorithm. Of course, the above if is a big if: it seems difficult to find closed form solutions when using a Metropolis-Hastings algorithm for instance and the paper only contains illustrations within the conjugate prior/Gibbs sampling framework. The second difficulty is also met by Dellaportas and Kontoyiannis, who show that the asymptotic variance of the resulting central limit can be equal to zero in some cases.