Archive for the Mountains Category

Guiana impressions [#2]

Posted in Kids, Mountains, pictures, Running, Travel with tags , , , , , , , , , , , , , , , , , , , on January 24, 2022 by xi'an

As in other oversea French territories, Guiana is a mix of reproduction of some metropolitan features (e.g., the postoffice, the gendarmerie, the signage, the large malls with French brands like Carrefour and the same products as here, the many boulangeries) and of local specificities or idiosyncrasies (e.g., very well maintained roadsides, haphazard garbage collection, convenience stores overwhelmingly operated by people of Chinese descent, hardly any regulation on guns, hunting, or bushmeat for non-protected species). Car wrecks are left along the roads, while the driving code there reminded me of the 80’s! Meaning risky overtakes, moppets with no lights and moppet drivers with no helmet. I also drove there possibly the worst ride of my life, over the 50km between Roura and the harbour on Kaw marshes, as the formally (or formerly?) paved D6 road is littered with potholes that are rarely avoidable and often quite deep. The drive back in the night, the rain and the fog was a nightmare!

We visited the launching site of ESA, Kourou, an impressive structure over a huge territory. But missed the James Webb launch by six hours, only catching the exhaust fumes of the rocket when we were approaching Cayenne (after a rather uncomfortable flight between a massive and man-spreading left neighbour and a reclining-to-the-max front neighbour). And missed a jaguar crossing the road by being in the “wrong” ESA bus! (Unless this is a usual line of the tour guide.)

As Amazon France does not truly work in French Amazonia (another idiosyncrasy!), for obvious cost and delay resaons, bookstores in cities like Cayenne and Kourou are terrific and hopefully standing a better chance of surviving. When we spent an hour in La Case à Bulles, the place was crowded! (As I forgot my regular Olympus camera at home, I would have loved to get Amazon delivery. Instead I bought a basic Kodak camera from the local supermarket, which returned most of these blah pictures before the batteries prematurely died.)

Guiana impressions [#1]

Posted in Books, Kids, Mountains, pictures, Running, Travel with tags , , , , , , , , , , , , , , , , , , , , , , , , on January 16, 2022 by xi'an

As our daughter Rachel has started her (five year) medical residency with a semester round in a French Guiana hospital, we took the opportunity of the Xmas break and of acceptable travel restrictions to visit her and the largest (and sole American) French departement for a week! This was a most unexpected trip that we enjoyed considerably.

While hot and humid is not my favourite type of weather (!) the weather remained quite tolerable that week, esp. when considering this was the start of the rain season (guiana means land of plentiful water in Arawak!) This made hiking on the (well-traced) paths in the local equatorial rain forest rather interesting, as the red soil is definitely muddy or worse. I however faced much less insects than I feared and mosquito bites were rare beyond the dawn and dusk periods. Plenty of birds, albeit mostly invisible. Except for the fantastic marshes of Kaw, where the variety of birds is amazing, including aras and toucans. Very muddy trails, did I mention it, but beautiful explosion of trees. Green everywhere.My first sight of a sloth was quite the treat, but I regret not spotting anteaters. Or a tapir. Swimming in the marshes of Kaw was great as well, with no worry from local caimans! Which we went spotting after nightfall. The place reminded me in several ways of Tonlé Sap lake, near Angkor.

Ate there an atipa bosco fish from the same place. Which has samurai armor. And two front legs to move outside water! As we had no say in what was served, we also ate paca meat in this restaurant, the agouti paca being a local rodent. Unfortunately because bush meat should not be served to tourists for fear of reducing the animal populations.

Visited several remains of former penal colonies, the whole country being a French penal colony at a not-so-distant-time, from the era when Louisiana was sold to the U.S. to the abolition in 1938, only implemented in 1953… Appalling to think that political and criminal prisoners were sent there to slowly rot to death, with no economical purpose on top of it! To the point of dead prisoners being immersed at sea rather than buried on island gallows, the local cemetery being reserved to guardians and their families….

agaves [jatp]

Posted in Mountains, pictures, Travel, Wines with tags , , , , , , , , on December 13, 2021 by xi'an

Cuajimoloyas [jatp]

Posted in Mountains, pictures, Running, Travel with tags , , , , , , , on December 11, 2021 by xi'an

21w5107 [day 2]

Posted in Books, Mountains, pictures, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , on December 1, 2021 by xi'an

After a rich and local (if freezing) dinner on a rooftop facing the baroque Oaxaca cathedral, and an early invigorating outdoor swim in my case!, the morning session was mostly on mixtures, with Helen Ogden exploring X validation for (estimating the number k of components for) finite mixtures, when using the likelihood as an objective function. I was unclear of the goal however when considering that the data supporting the study was Uniform (0,1), nothing like a mixture of Normal distributions. And about the consistency attached to the objective function. The session ended with Diana Cai presenting a counter-argument in the sense that she proved, along with Trevor Campbell and Tamara Broderick, that the posterior on k diverges to infinity with the number n of observations if a mixture model is misspecified for said data. Which does not come as a major surprise since there is no properly defined value of k when the data is not generated from the adopted mixture. I would love to see an extension to the case when the k component mixture contains a non-parametric component! In-between, Alexander Ly discussed Bayes factors for multiple datasets, with some asymptotics showing consistency for some (improper!) priors if one sample size grows to infinity. With actually attaining the same rate under both hypotheses. Luis Nieto-Barajas presented an approach on uncertainty assessment through KL divergence for random probability measures, which requires a calibration of the KL in this setting, as KL does not enjoy a uniform scale, and a prior on a Pólya tree. And Chris Holmes presented a recent work with Edwin Fong and Steven Walker on a prediction approach to Bayesian inference. Which I had had on my reading list for a while. It is a very original proposal where likelihoods and priors are replaced by the sequence of posterior predictives and only parameters of interest get simulated. The Bayesian flavour of the approach is delicate to assess though, albeit a form of non-parametric Bayesian perspective… (I still need to read the paper carefully.)

In the afternoon session, Judith Rousseau presented her recent foray in cut posteriors for semi-parametric HMMs. With interesting outcomes for efficiently estimating the transition matrix, the component distributions, and the smoothing distribution. I wonder at the connection with safe Bayes in that cut posteriors induce a loss of information. Sinead Williamson spoke on distributed MCMC for BNP. Going back at the “theme of the day”, namely clustering and finding the correct (?) number of clusters. With a collapsed versus uncollapsed division that reminded me of the marginal vs. conditional María Gil-Leyva discussed yesterday. Plus a decomposition of a random measure into a finite mixture and an infinite one that also reminded me of the morning talk of Diana Cai. (And making me wonder at the choice of the number K of terms in the finite part.) Michele Guindani spoke about clustering distributions (with firecrackers as a background!). Using the nDP mixture model, which was show to suffer from degeneracy (as discussed by Frederico Camerlenghi et al. in BA). The subtle difference stands in using the same (common) atoms in all random distributions at the top of the hierarchy, with independent weights. Making the partitions partially exchangeable. The approach relies on Sylvia’s generalised mixtures of finite mixtures. With interesting applications to microbiome and calcium imaging (including a mice brain in action!). And Giovanni Rebaudo presented a generalised notion of clustering aligned on a graph, with some observations located between the nodes corresponding to clusters. Represented as a random measure with common parameters for the clusters and separated parameters outside. Interestingly playing on random partitions, Pólya urns, and species sampling.

%d bloggers like this: