Archive for the Running Category

a journal of the no-end-war year

Posted in Books, Kids, Mountains, pictures, Running, Travel, University life, Wines with tags , , , , , , , , , , , , , , , , , , , , , , , , , , , , , on May 6, 2026 by xi'an

Read my first book in Italian!, namely the fumetto Favola di Venezia (Sirat Al-Bunduqiyyah) by Hugo Pratt. Which I bought in the nice bookstore of the Santa Lucia train station rather than in the new store dedicated to Corto Maltese, in the Ghetto. It took me a while, as I had to get through word by word, rather than rushing through the dialogues as in French BDs or US comics. But this was most appropriate for a fairly oniric story that is more an homage to Pratt’s childhood place than a political mystery (even though the growth of fascism is also impacting the story). Going through each block that slowly further made me try to recognise some places (without checking websites that identify each real location!). As a coïncidence, The New York Times published a long review of… Fable of Venice and Other Adventures (other meaning “The Secret of Tristan Bantam,” “So Much for Gentlemen of Fortune,” and “The Seagull’s Fault,”) which goes over Pratt’s childhood, in Venice, his family (a grand-father joining early the fascists, via Mussolini’s Fasci Italiani di Combattimento, a father moving the family to Ethiopia in 1937 as an overseer of a fascist labour camp, later dying in an Allied prisoner camp),  his adventures as an errand boy in an Ethiopian brothel, then as a translator for the Allies occupying Venice, and later moving to Buenos Aires to work for a publisher… (Incidentally, I realised during that trip that Milano and Torino only got liberated on 25 Aprile 1945, a few days before the official end of WW II in Europe. Venezia and Padua got liberated on 28 Aprile.)

During my last trip (this semester!) in Venezia, I was (almost) kidnapped for a dinner at la Trattoria alle Vignole, located on the Isola le Vignole, which I had never visited before. (It is located east of Murano, facing the entrance to the Arsenale.) Which proved exceptional in many ways, from the isola(tion), to the absence of any other customers, to the ferry ride, to watching the sunset on Cannaregio and Murano, and of course to the endless flow of sea-food and fish dishes (not mentioning the company!!)… After tasting (or rather devouring!) a colomba pasquale (for Easter), I also brought home several versions of the similar foccacia veneta, for a battle of bakers!

Watched Phantom Lawyer, a rather silly, heavy going, Korean TV series featuring a struggling lawyer who can see ghosts and bring their cases to a satisfactory conclusion. With the usual conflicts of interest, massive coïncidences, and hopelessly slow romantic entanglements.

inverse probability weighting

Posted in Books, pictures, Running, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , on May 4, 2026 by xi'an

Quite recently, Jyotishka Datta and Nick Polson published a fairly interesting [imho] paper in The New England Journal of Statistics in Data Science, entitled Inverse Probability Weighting: From Survey Sampling to Evidence Estimation that (obviously) caters to my own interests! They bring three threads together. First, they recall the long debate between using [normalized] Horvitz–Thompson and [self-normalized] Hájek estimators in survey sampling, pointing out that the latter is “usually the better estimator, despite estimation of an a priori known quantity” (citing from Särndal & al., 2003). Which is also my experience with importance sampling, as in this 1995 Note aux Comptes Rendus with George. The mathematical paradox of “estimating” a constant is central to other advances in the area, like noise-contrastive estimation à la Gutmann & Hyvärinen (2005) or the measure estimation of Kong & al. (2003). Even more interestingly, Datta & Polson consider there is a link with the inconsistent Bayesian (counter)example of Larry Wasserman and Jamie Robins, where the censoring probability increases with the value of the parameter of interest, paradox in which Chris Sims also got involved. (I was unaware that he had passed away last month.) And, lo and behold!, with the Stein “paradox” of my PhD years (and beyond).

Their central argument stands with the missing data link between Horvitz-Thompson survey sampling and Monte Carlo integration. (With a reference to our Riemann sum papers with Anne Philippe!, making me realise the authors had recently published an extension on that idea.) This reminds me very much of the missing measure approach of Kong et al. (2003). (Actually the reference appears in the final discussion.) The authors go over several paradoxes like Basu’s circus estimate (1988), Larry’s inconsistent Bayes estimate (2004), where Horvitz–Thompson performs nicely under compactness assumptions, the Bayesian answers (which include nested sampling even though I do not see the connection). Especially Li’s (2010) solution.  The attached numerical experiment displays a consistent underperformance of the Horvitz-Thompson estimator, in contrast with the theory… 

In conclusion, while enjoying very much revisiting so many examples and papers I came across in the past decades, I remain somewhat puzzled by the lack of overall message.

 

un.be.lie.va.ble! [thrice]

Posted in Running with tags , , , , , , , , , , on April 28, 2026 by xi'an

Straßburger Abenrot

Posted in pictures, Running, Travel, University life with tags , , , , , , , , , , , , , , , , , on April 27, 2026 by xi'an

settimana scorsa a Venezia

Posted in pictures, Running, Travel, University life with tags , , , , , , , , , , , , , , , , , , on April 21, 2026 by xi'an