The next mostly MCMC seminar is actually happening today and (darn!) I forgot to advertise it! For those who can still make it, it will take place at 3PM (CEST) at the PariSanté Campus (in Room 08). The talks are by
After a long absence from the monthly Séminaire Parisien de Statistique I attended one today at IHP, including a talk by Diala Hawat on repelled point processes for numerical integration by Hawat et al. The goal is to get (and prove) a universal variance improvement for numerical integration by applying a form of determinantal processes to initial simulations, as eg iid (Poisson process) sampling (without accounting for the O(N²) cost in moving these points). The repelled points are obtain by a single (why single?) move based on a force function (as shown in the slide below), inspired by a Coulomb potential (in the sense that said move appears as one gradient step along the potential). Which reminded me of the pinball sampler, even though the inverse norm was just there to create infinite repulsion near each point. A surprising feature of this repelling step is that it even modifies a (QMC) Sobol process with also an (empirical) improvement in the variance. I wonder if one could construct an MCMC algorithm that would target a joint distribution, maybe via a copula representation, maybe via an equivalent version of HMC.
As an aside, the Bakhvalov results on the existence of a worst case integrand for any deterministic or random sequence (see top slide) made me wonder what the shape of this worst case function is, esp. for a QMC sequence (eg, Sobol). And whether or not they are of any relevance as a counterfactor to the optimal importance functions.
On Monday, I took part in a celebration of the remarkable career of a former colleague of mine in Rouen, Gérard Grancher, who is retiring after a life-long position as CNRS engineer in the department of maths of the University of Rouen, a job title that tells very little about the numerous facets of his interactions with mathematics, from his handling of all informatics aspects in the laboratory to his support of all colleagues there, including fresh PhD students like me in 1985!, to his direction of the CNRS lab in 2006 and 2007 at a time of deep division and mistrust, to his numerous collaborations on statistical projects with local actors, to his Norman federalism in bringing the maths departments of Caen and Rouen into a regional federation, to an unceasing activism to promote maths in colleges and high schools and science fairs all around Normandy, to his contributions to professional training in statistics for CNRS agents, and much, much more… Which explains why the science auditorium of the University of Rouen was packed with mathematicians and high schools maths teachers and friends! (The poster of the day was made by Gérard’s accomplices in vulgarisation, Élise Janvresse and Thierry Delarue, based on a sample of points randomly drawn from Gérard’s picture, maybe using a determinantal process, and the construction of a travelling salesman path over those points.)
This was a great day with mostly vulgarisation talks (including one about Rasmus’ socks..!) and reminiscences about Gérard’s carreer at Rouen. As I had left the university in 2000 to move to Paris-Dauphine, this was a moving day as well, as I met with old friends I had not seen for ages, including our common PhD advisor, Jean-Pierre Raoult.
This trip back in time was also an opportunity to (re-)visit the beautifully preserved medieval centre of Rouen, with its wooden houses, Norman-style, the numerous churches, including Monet‘s cathedral, the Justice Hall… Last time I strolled those streets, George Casella was visiting!
Shinichiro Shirota and Alan Gelfand arXived a paper on the use of ABC for analysing some repulsive point processes, more exactly the Gibbs point processes, for which ABC requires a perfect sampler to operate, unless one is okay with stopping an MCMC chain before it converges, and determinantal point processes studied by Lavancier et al. (2015) [a paper I wanted to review and could not find time to!]. Detrimental point processes have an intensity function that is the determinant of a covariance kernel, hence repulsive. Simulation of a determinantal process itself is not straightforward and involves approximations. But the likelihood itself is unavailable and Lavancier et al. (2015) use approximate versions by fast Fourier transforms, which means MCMC is challenging even with those approximate steps.
“The main computational cost of our algorithm is simulation of x for each iteration of the ABC-MCMC.”
The authors propose here to use ABC instead. With an extra approximative step for simulating the determinantal process itself. Interestingly, the Gibbs point process allows for a sufficient statistic, the number of R-closed points, although I fail to see how the radius R is determined by the model, while the determinantal process does not. The summary statistics end up being a collection of frequencies within various spheres of different radii. However, these statistics are then processed by Fearnhead’s and Prangle’s proposal, namely to come up as an approximation of E[θ|y] as the natural summary. Obtained by regression over the original summaries. Another layer of complexity stems from using an ABC-MCMC approach. And including a Lasso step in the regression towards excluding less relevant radii. The paper also considers Bayesian model validation for such point processes, implementing prior predictive tests with a ranked probability score, rather than a Bayes factor.
As point processes have always been somewhat mysterious to me, I do not have any intuition about the strength of the distributional assumptions there and the relevance of picking a determinantal process against, say, a Strauss process. The model comparisons operated in the paper are not strongly supporting one repulsive model versus the others, with the authors concluding at the need for many points towards a discrimination between models. I also wonder at the possibility of including other summaries than Ripley’s K-functions, which somewhat imply a discretisation of the space, by concentric rings. Maybe using other point processes for deriving summary statistics as MLEs or Bayes estimators for those models would help. (Or maybe not.)
 | David on Bayesian workflow [book review… |
 | Ewan on Tidy up, Claude! |
 | xi'an on Tidy up, Claude! |
| Ewan on Tidy up, Claude! |
| Ewan on Já við viðræðum. Já við Evrópu… |
